AI For Business Microsoft

AI For Business Microsoft — independent reviews, comparisons, pricing and step-by-step guides on Aizhi.

• 

  DaVinci (software)

  DaVinci was a development tool produced by Incross, which aimed at creating HTML5 mobile applications and media content. It included a jQuery framework and a JavaScript library that enabled developers and designers to craft web applications designed for mobile devices with a user experience similar to native applications. Business applications, games, rich media content, such as HTML5 multi-media magazines, advertisements, and animation, may be produced with the tool. DaVinci was based on standard web technology – including HTML5, CSS3, and JavaScript. == Features == DaVinci comprised DaVinci Studio and DaVinci Animator, which handled application programming and UI design. The tool had a WYSIWYG authoring environment. Open-source libraries, such as KnockOut, JsRender/JsViews, Impress.js, and turn.js, were included in the tool. Other open-source frameworks could also be integrated. The Model View Controller (MVC) and Data Binding in JavaScript could be handled through DaVinci's Data-Set Editor. In this mode, view components and model data could be visually bound, which allowed users to create web applications with server-integrated UI components without coding. Additionally, DaVinci included an N-Screen editor, which automatically adjusted designs and functionalities to fit the screen sizes of various devices, including smartphones, tablet PCs, and TVs. == DaVinci and jQuery == In collaboration with the jQuery Foundation, DaVinci played a significant role in hosting the first jQuery conference in an Asian district, which took place on November 12, 2012, in Seoul, South Korea. The conference showcased how DaVinci could be utilized in application development demonstrations.

  Read more →
• 

  Jerome H. Friedman

  Jerome Harold Friedman (born December 29, 1939) is an American statistician, consultant and Professor of Statistics at Stanford University, known for his contributions in the field of statistics and data mining. == Biography == Friedman studied at Chico State College for two years before transferring to the University of California, Berkeley in 1959, where he received his AB in Physics in 1962, and his PhD in High Energy Particle Physics in 1967. In 1968 he started his academic career as research physicist at the Lawrence Berkeley National Laboratory. In 1972 he started at Stanford University as leader of the Computation Research Group at the Stanford Linear Accelerator Center, where he would participate until 2003. In the year 1976–77 he was a visiting scientist at CERN in Geneva. From 1981 to 1984 he was visiting professor at the University of California, Berkeley. In 1982 he was appointed Professor of Statistics at Stanford University. In 1984 he was elected as a Fellow of the American Statistical Association. In 2002 he was awarded the SIGKDD Innovation Award by the Association for Computing Machinery (ACM). In 2010 he was elected as a member of the National Academy of Sciences (Applied mathematical sciences). == Publications == Friedman has authored and co-authored many publications in the field of data-mining including "nearest neighbor classification, logistical regressions, and high dimensional data analysis. His primary research interest is in the area of machine learning." A selection: Friedman, Jerome H. & Tukey, John W. (1974). "A projection pursuit algorithm for exploratory data analysis". IEEE Transactions on Computers. 23 (9): 881–890. doi:10.1109/T-C.1974.224051. OSTI 1442925. S2CID 7997450. Friedman, Jerome H. & Stuetzle, Werner (1981). "Projection pursuit regression". Journal of the American Statistical Association. 76 (376): 817–823. doi:10.1080/01621459.1981.10477729. OSTI 1445517. Friedman, Jerome H. (1991). "Multivariate adaptive regression splines". Annals of Statistics. 19 (1): 1–67. CiteSeerX 10.1.1.382.970. doi:10.1214/aos/1176347963. JSTOR 2241837. Friedman, Jerome H. (2001). "Greedy function approximation: a gradient boosting machine". Annals of Statistics. 29 (5): 1189–1232. doi:10.1214/aos/1013203451. JSTOR 2699986.

  Read more →
• 

  Sasha Luccioni

  Alexandra Sasha Luccioni (née Vorobyova; born 1990) is a computer scientist specializing in the intersection of artificial intelligence (AI) and climate change. Her work focuses on quantifying the environmental impact of AI technologies and promoting sustainable practices in machine learning development. == Early life and education == Alexandra Sasha Vorobyova was born in the Ukrainian Soviet Socialist Republic in 1990. When she was four years old, her family relocated to Ontario, Canada. Her interest in science is influenced by her family's history; her mother, grandmother, and great-grandmother all pursued careers in scientific fields. Luccioni earned a B.A. in language science from University of Paris III: Sorbonne Nouvelle in 2010. Subsequently, she completed a M.S. in cognitive science, with a minor in natural language processing, at École normale supérieure in Paris in 2012. Luccioni obtained her PhD in cognitive computing from Université du Québec à Montréal (UQAM) in 2018. == Career == Luccioni began her professional career at Nuance Communications in 2017, where she focused on natural language processing (NLP) and machine learning (ML) techniques to enhance conversational agents. She then joined Morgan Stanley’s AI/ML Center of Excellence in 2018, working on explainable artificial intelligence (AI) and decision-making systems. In 2019, she became a postdoctoral researcher at Université de Montréal and Mila, collaborating with computer scientist Yoshua Bengio on a project titled This Climate Does Not Exist. This initiative used generative adversarial networks to visualize the effects of climate change. During this time, she also contributed to integrating fairness and accountability into machine learning education at Mila. Luccioni briefly worked with the United Nations Global Pulse in 2021, developing tools to monitor COVID-19 misinformation. Later that year, she joined Hugging Face as a research scientist. Her role includes quantifying the carbon footprint of AI systems, co-chairing the carbon working group in the Big Science project, and advancing responsible machine learning practices. She helped create "CodeCarbon," an open-source software tool that estimates the carbon emissions produced during the training and operation of machine learning models. In addition to her research, she has developed tools to measure the environmental impact of AI models, communicated findings through media engagements, and presented at international conferences, including a TED Talk. In 2024, she was listed on BBC 100 Women and Time 100 AI.

  Read more →
• 

  How to Choose an AI Bug Finder

  Comparing the best AI bug finder? An AI bug finder is software that uses machine learning to help you get more done — it lowers the barrier so anyone can produce professional output. Privacy matters too: check whether your data trains the model and whether a no-log or enterprise tier is available. Whether you are a beginner or a pro, the right AI bug finder slots into your workflow and pays for itself fast. Below we compare features, pricing, and real output so you can choose with confidence.

  Read more →
• 

  Cloud testing

  Cloud testing is a form of software testing in which web applications use cloud computing environments (a "cloud") to simulate real-world user traffic. == Steps == Companies simulate real world Web users by using cloud testing services that are provided by cloud service vendors such as Advaltis, Compuware, HP, Keynote Systems, Neotys, RadView and SOASTA. Once user scenarios are developed and the test is designed, these service providers leverage cloud servers (provided by cloud platform vendors such as Amazon.com, Google, Rackspace, Microsoft, etc.) to generate web traffic that originates from around the world. Once the test is complete, the cloud service providers deliver results and analytics back to corporate IT professionals through real-time dashboards for a complete analysis of how their applications and the internet will perform during peak volumes. == Applications == Cloud testing is often seen as only performance or load tests, however, as discussed earlier it covers many other types of testing. Cloud computing itself is often referred to as the marriage of software as a service (SaaS) and utility computing. In regard to test execution, the software offered as a service may be a transaction generator and the cloud provider's infrastructure software, or may just be the latter. Distributed Systems and Parallel Systems mainly use this approach for testing, because of their inherent complex nature. D-Cloud is an example of such a software testing environment. == Tools == Leading cloud computing service providers include, among others, Amazon, Microsoft, Google, RadView, Skytap, HP and SOASTA. == Benefits == The ability and cost to simulate web traffic for software testing purposes has been an inhibitor to overall web reliability. The low cost and accessibility of the cloud's extremely large computing resources provides the ability to replicate real world usage of these systems by geographically distributed users, executing wide varieties of user scenarios, at scales previously unattainable in traditional testing environments. Minimal start-up time along with quality assurance can be achieved by cloud testing. Following are some of the key benefits: Reduction in capital expenditure Highly scalable

  Read more →
• 

  Alexei A. Efros

  Alexei "Alyosha" A. Efros (born 9 April 1975) is a Russian-American computer scientist and professor at University of California, Berkeley. He has contributed to the field of computer vision, and his work has been referenced in Wired, BBC News, The New York Times, and The New Yorker. == Early life and education == Efros was born in St. Petersburg in the Soviet Union. His father is Alexei L. Efros, then a physics professor at the Ioffe Physico-Technical Institute. His family emigrated to the United States when he was 14 to accommodate his father's career and the family settled in Salt Lake City in 1991. He graduated from the University of Utah in 1997, and attended University of California, Berkeley for his PhD, where he was advised by Jitendra Malik and graduated in 2003. He then spent a year as a research fellow at the University of Oxford, where he worked with Andrew Zisserman. == Career == Efros joined the faculty at Carnegie Mellon University in Pittsburgh, where he remained until 2013 when he joined the faculty of the University of California, Berkeley. He received a Guggenheim Fellowship in 2008. He received the 2016 ACM Prize in Computing.

  Read more →
• 

  Synchronizing word

  In computer science, more precisely, in the theory of deterministic finite automata (DFA), a synchronizing word or reset sequence is a word in the input alphabet of the DFA that sends any state of the DFA to one and the same state. That is, if an ensemble of copies of the DFA are each started in different states, and all of the copies process the synchronizing word, they will all end up in the same state. Not every DFA has a synchronizing word; for instance, a DFA with two states, one for words of even length and one for words of odd length, can never be synchronized. == Existence == Given a DFA, the problem of determining if it has a synchronizing word can be solved in polynomial time using a theorem due to Ján Černý. A simple approach considers the power set of states of the DFA, and builds a directed graph where nodes belong to the power set, and a directed edge describes the action of the transition function. A path from the node of all states to a singleton state shows the existence of a synchronizing word. This algorithm is exponential in the number of states. A polynomial algorithm results however, due to a theorem of Černý that exploits the substructure of the problem, and shows that a synchronizing word exists if and only if every pair of states has a synchronizing word. == Length == The problem of estimating the length of synchronizing words has a long history and was posed independently by several authors, but it is commonly known as the Černý conjecture. In 1969, Ján Černý conjectured that (n − 1)2 is the upper bound for the length of the shortest synchronizing word for any n-state complete DFA (a DFA with complete state transition graph). If this is true, it would be tight: in his 1964 paper, Černý exhibited a class of automata (indexed by the number n of states) for which the shortest reset words have this length. The best upper bound known is 0.1654n3, far from the lower bound. For n-state DFAs over a k-letter input alphabet, an algorithm by David Eppstein finds a synchronizing word of length at most 11n3/48 + O(n2), and runs in time complexity O(n3+kn2). This algorithm does not always find the shortest possible synchronizing word for a given automaton; as Eppstein also shows, the problem of finding the shortest synchronizing word is NP-complete. However, for a special class of automata in which all state transitions preserve the cyclic order of the states, he describes a different algorithm with time O(kn2) that always finds the shortest synchronizing word, proves that these automata always have a synchronizing word of length at most (n − 1)2 (the bound given in Černý's conjecture), and exhibits examples of automata with this special form whose shortest synchronizing word has length exactly (n − 1)2. == Road coloring == The road coloring problem is the problem of labeling the edges of a regular directed graph with the symbols of a k-letter input alphabet (where k is the outdegree of each vertex) in order to form a synchronizable DFA. It was conjectured in 1970 by Benjamin Weiss and Roy Adler that any strongly connected and aperiodic regular digraph can be labeled in this way; their conjecture was proven in 2007 by Avraham Trahtman. == Related: transformation semigroups == A transformation semigroup is synchronizing if it contains an element of rank 1, that is, an element whose image is of cardinality 1. A DFA corresponds to a transformation semigroup with a distinguished generator set.

  Read more →
• 

  Structured prediction

  Structured prediction or structured output learning is an umbrella term for supervised machine learning techniques that involves predicting structured objects, rather than discrete or real values. Similar to commonly used supervised learning techniques, structured prediction models are typically trained by means of observed data in which the predicted value is compared to the ground truth, and this is used to adjust the model parameters. Due to the complexity of the model and the interrelations of predicted variables, the processes of model training and inference are often computationally infeasible, so approximate inference and learning methods are used. == Applications == An example application is the problem of translating a natural language sentence into a syntactic representation such as a parse tree. This can be seen as a structured prediction problem in which the structured output domain is the set of all possible parse trees. Structured prediction is used in a wide variety of domains including bioinformatics, natural language processing (NLP), speech recognition, and computer vision. === Example: sequence tagging === Sequence tagging is a class of problems prevalent in NLP in which input data are often sequential, for instance sentences of text. The sequence tagging problem appears in several guises, such as part-of-speech tagging (POS tagging) and named entity recognition. In POS tagging, for example, each word in a sequence must be 'tagged' with a class label representing the type of word: The main challenge of this problem is to resolve ambiguity: in the above example, the words "sentence" and "tagged" in English can also be verbs. While this problem can be solved by simply performing classification of individual tokens, this approach does not take into account the empirical fact that tags do not occur independently; instead, each tag displays a strong conditional dependence on the tag of the previous word. This fact can be exploited in a sequence model such as a hidden Markov model or conditional random field that predicts the entire tag sequence for a sentence (rather than just individual tags) via the Viterbi algorithm. == Techniques == Probabilistic graphical models form a large class of structured prediction models. In particular, Bayesian networks and random fields are popular. Other algorithms and models for structured prediction include inductive logic programming, case-based reasoning, structured SVMs, Markov logic networks, Probabilistic Soft Logic, and constrained conditional models. The main techniques are: Conditional random fields Structured support vector machines Structured k-nearest neighbours Recurrent neural networks, in particular Elman networks Transformers. === Structured perceptron === One of the easiest ways to understand algorithms for general structured prediction is the structured perceptron by Collins. This algorithm combines the perceptron algorithm for learning linear classifiers with an inference algorithm (classically the Viterbi algorithm when used on sequence data) and can be described abstractly as follows: First, define a function ϕ ( x , y ) {\displaystyle \phi (x,y)} that maps a training sample x {\displaystyle x} and a candidate prediction y {\displaystyle y} to a vector of length n {\displaystyle n} ( x {\displaystyle x} and y {\displaystyle y} may have any structure; n {\displaystyle n} is problem-dependent, but must be fixed for each model). Let G E N {\displaystyle GEN} be a function that generates candidate predictions. Then: Let w {\displaystyle w} be a weight vector of length n {\displaystyle n} For a predetermined number of iterations: For each sample x {\displaystyle x} in the training set with true output t {\displaystyle t} : Make a prediction y ^ {\displaystyle {\hat {y}}} : y ^ = a r g m a x { y ∈ G E N ( x ) } ( w T , ϕ ( x , y ) ) {\displaystyle {\hat {y}}={\operatorname {arg\,max} }\,\{y\in GEN(x)\}\,(w^{T},\phi (x,y))} Update w {\displaystyle w} (from y ^ {\displaystyle {\hat {y}}} towards t {\displaystyle t} ): w = w + c ( − ϕ ( x , y ^ ) + ϕ ( x , t ) ) {\displaystyle w=w+c(-\phi (x,{\hat {y}})+\phi (x,t))} , where c {\displaystyle c} is the learning rate. In practice, finding the argmax over G E N ( x ) {\displaystyle {GEN}({x})} is done using an algorithm such as Viterbi or a max-sum, rather than an exhaustive search through an exponentially large set of candidates. The idea of learning is similar to that for multiclass perceptrons.

  Read more →
• 

  Situational application

  In computing, a situational application is "good enough" software created for a narrow group of users with a unique set of needs. The application typically (but not always) has a short life span, and is often created within the group where it is used, sometimes by the users themselves. As the requirements of a small team using the application change, the situational application often also continues to evolve to accommodate these changes. Although situational applications are specifically designed to embrace change, significant changes in requirements may lead to an abandonment of the situational application altogether – in some cases it is just easier to develop a new one than to evolve the one in use. == Characteristics == Situational applications are developed fast, easy to use, uncomplicated, and serve a unique set of requirements. They have a narrow focus on a specific business problem, and they are written in a way where if the business problem changes rapidly, so can the situational application. This contrasts with more common enterprise applications, which are designed to address a large set of business problems, require meticulous planning, and impose a sometimes-slow and often-meticulous change process. == Origination == Clay Shirky in his essay entitled "Situated Software" described a type of software that "...is designed for use by a specific social group, rather than for a generic set of "users"." IBM later morphed the term into "situational applications". == Evolution == The successful large-scale implementation of a situational application environment in an organization requires a strategy, mindset, methodology and support structure quite different from traditional application development. This is now evolving as more companies learn how to best leverage the ideas behind situational applications. In addition, the advent of cloud-based application development and deployment platforms makes the implementation of a comprehensive situational application environment much more feasible. == Examples == A structured wiki that can host wiki applications lends itself to creation of situational applications. Some mashups can also be considered situational applications. A forms application such as a Microsoft Access Database (MDB file) can be considered a situational application. The latest implementations of situational application environments include Longjump, Force.com and WorkXpress.

  Read more →
• 

  Statistical machine translation

  Statistical machine translation (SMT) is a machine translation approach where translations are generated on the basis of statistical models whose parameters are derived from the analysis of bilingual text corpora. The statistical approach contrasts with the rule-based approaches to machine translation as well as with example-based machine translation, that superseded the previous rule-based approach that required explicit description of each and every linguistic rule, which was costly, and which often did not generalize to other languages. The first ideas of statistical machine translation were introduced by Warren Weaver in 1949, including the ideas of applying Claude Shannon's information theory. Statistical machine translation was re-introduced in the late 1980s and early 1990s by researchers at IBM's Thomas J. Watson Research Center. Before the introduction of neural machine translation, it was by far the most widely studied machine translation method. == Basis == The idea behind statistical machine translation comes from information theory. A document is translated according to the probability distribution p ( e | f ) {\displaystyle p(e|f)} that a string e {\displaystyle e} in the target language (for example, English) is the translation of a string f {\displaystyle f} in the source language (for example, French). The problem of modeling the probability distribution p ( e | f ) {\displaystyle p(e|f)} has been approached in a number of ways. One approach which lends itself well to computer implementation is to apply Bayes' theorem, that is p ( e | f ) ∝ p ( f | e ) p ( e ) {\displaystyle p(e|f)\propto p(f|e)p(e)} , where the translation model p ( f | e ) {\displaystyle p(f|e)} is the probability that the source string is the translation of the target string, and the language model p ( e ) {\displaystyle p(e)} is the probability of seeing that target language string. This decomposition is attractive as it splits the problem into two subproblems. Finding the best translation e ~ {\displaystyle {\tilde {e}}} is done by picking up the one that gives the highest probability: e ~ = a r g max e ∈ e ∗ p ( e | f ) = a r g max e ∈ e ∗ p ( f | e ) p ( e ) {\displaystyle {\tilde {e}}=arg\max _{e\in e^{}}p(e|f)=arg\max _{e\in e^{}}p(f|e)p(e)} . For a rigorous implementation of this one would have to perform an exhaustive search by going through all strings e ∗ {\displaystyle e^{}} in the native language. Performing the search efficiently is the work of a machine translation decoder that uses the foreign string, heuristics and other methods to limit the search space and at the same time keeping acceptable quality. This trade-off between quality and time usage can also be found in speech recognition. As the translation systems are not able to store all native strings and their translations, a document is typically translated sentence by sentence. Language models are typically approximated by smoothed n-gram models, and similar approaches have been applied to translation models, but this introduces additional complexity due to different sentence lengths and word orders in the languages. Statistical translation models were initially word based (Models 1-5 from IBM Hidden Markov model from Stephan Vogel and Model 6 from Franz-Joseph Och), but significant advances were made with the introduction of phrase based models. Later work incorporated syntax or quasi-syntactic structures. == Benefits == The most frequently cited benefits of statistical machine translation (SMT) over rule-based approach are: More efficient use of human and data resources There are many parallel corpora in machine-readable format and even more monolingual data. Generally, SMT systems are not tailored to any specific pair of languages. More fluent translations owing to use of a language model == Shortcomings == Corpus creation can be costly. Specific errors are hard to predict and fix. Results may have superficial fluency that masks translation problems. Statistical machine translation usually works less well for language pairs with significantly different word order. The benefits obtained for translation between Western European languages are not representative of results for other language pairs, owing to smaller training corpora and greater grammatical differences. == Word-based translation == In word-based translation, the fundamental unit of translation is a word in some natural language. Typically, the number of words in translated sentences are different, because of compound words, morphology and idioms. The ratio of the lengths of sequences of translated words is called fertility, which tells how many foreign words each native word produces. Necessarily it is assumed by information theory that each covers the same concept. In practice this is not really true. For example, the English word corner can be translated in Spanish by either rincón or esquina, depending on whether it is to mean its internal or external angle. Simple word-based translation cannot translate between languages with different fertility. Word-based translation systems can relatively simply be made to cope with high fertility, such that they could map a single word to multiple words, but not the other way about. For example, if we were translating from English to French, each word in English could produce any number of French words— sometimes none at all. But there is no way to group two English words producing a single French word. An example of a word-based translation system is the freely available GIZA++ package (GPLed), which includes the training program for IBM models and HMM model and Model 6. The word-based translation is not widely used today; phrase-based systems are more common. Most phrase-based systems are still using GIZA++ to align the corpus. The alignments are used to extract phrases or deduce syntax rules. And matching words in bi-text is still a problem actively discussed in the community. Because of the predominance of GIZA++, there are now several distributed implementations of it online. == Phrase-based translation == In phrase-based translation, the aim is to reduce the restrictions of word-based translation by translating whole sequences of words, where the lengths may differ. The sequences of words are called blocks or phrases. These are typically not linguistic phrases, but phrasemes that were found using statistical methods from corpora. It has been shown that restricting the phrases to linguistic phrases (syntactically motivated groups of words, see syntactic categories) decreased the quality of translation. The chosen phrases are further mapped one-to-one based on a phrase translation table, and may be reordered. This table could be learnt based on word-alignment, or directly from a parallel corpus. The second model is trained using the expectation maximization algorithm, similarly to the word-based IBM model. == Syntax-based translation == Syntax-based translation is based on the idea of translating syntactic units, rather than single words or strings of words (as in phrase-based MT), i.e. (partial) parse trees of sentences/utterances. Until the 1990s, with advent of strong stochastic parsers, the statistical counterpart of the old idea of syntax-based translation did not take off. Examples of this approach include DOP-based MT and later synchronous context-free grammars. == Hierarchical phrase-based translation == Hierarchical phrase-based translation combines the phrase-based and syntax-based approaches to translation. It uses synchronous context-free grammar rules, but the grammars can be constructed by an extension of methods for phrase-based translation without reference to linguistically motivated syntactic constituents. This idea was first introduced in Chiang's Hiero system (2005). == Language models == A language model is an essential component of any statistical machine translation system, which aids in making the translation as fluent as possible. It is a function that takes a translated sentence and returns the probability of it being said by a native speaker. A good language model will for example assign a higher probability to the sentence "the house is small" than to "small the is house". Other than word order, language models may also help with word choice: if a foreign word has multiple possible translations, these functions may give better probabilities for certain translations in specific contexts in the target language. == Systems implementing statistical machine translation == Google Translate (started transition to neural machine translation in 2016) Microsoft Translator (started transition to neural machine translation in 2016) Yandex.Translate (switched to hybrid approach incorporating neural machine translation in 2017) == Challenges with statistical machine translation == Problems with statistical machine translation include: === Sentence alignment === Single sentences in one language can be found translated into several sentences in the o

  Read more →
• 

  Luca Maria Gambardella

  Luca Maria Gambardella (born 4 January 1962) is an Italian computer scientist and author. He is the former director of the Dalle Molle Institute for Artificial Intelligence Research in Lugano, in the Ticino canton of Switzerland. He is currently the prorector of Università della Svizzera italiana, where he directs the Master of Science in Artificial Intelligence degree course. Several of his papers have been extensively cited, with his collaborators including Marco Dorigo, with whom he has published papers on the application of ant colony optimization theory to the traveling salesman problem, and Jürgen Schmidhuber with whom he has published research on deep neural networks.. Beside working in research, Gambardella explores the potentials of AI applied for the generation of art. Some of his artistic installations received significant media coverage. As a novelist, the genres he approached broad from Bildungsroman of his first book "Sei vite" ("Six lives"), to romance of his second book "Il suono dell'alba" ("The sound of sunrise").

  Read more →
• 

  Anil K. Jain (computer scientist, born 1948)

  Anil Kumar Jain (born 1948) is an Indian-American computer scientist and University Distinguished Professor in the Department of Computer Science and Engineering at Michigan State University. He is one of the most highly cited researchers in computer science, and is internationally recognized for his foundational contributions to pattern recognition, computer vision, and biometric recognition, particularly in fingerprint recognition and face recognition. Jain is a member of the United States National Academy of Engineering, a Foreign Member of the Chinese Academy of Sciences, and a Foreign Fellow of the Indian National Academy of Engineering. He is a Fellow of the ACM, IEEE, AAAS, IAPR, and SPIE. His research has shaped the field of biometrics and has been applied in systems used worldwide for identity verification, law enforcement, and border security. In 2024, he was awarded the BBVA Foundation Frontiers of Knowledge Award in the category of Information and Communication Technologies. == Early life and education == Born in Basti, India, Jain received his Bachelor of Technology in electrical engineering from the Indian Institute of Technology, Kanpur in 1969. He then moved to the United States, where he earned his M.S. in 1970 and Ph.D. in 1973 from Ohio State University. His doctoral dissertation, titled Some Aspects of Dimensionality and Sample Size Problems in Statistical Pattern Recognition, was supervised by Robert B. McGhee and laid the groundwork for his subsequent research in pattern recognition. == Career == Jain began his academic career at Wayne State University, where he taught from 1972 to 1974. In 1974, he joined the faculty of Michigan State University, where he has remained for over five decades and currently holds the position of University Distinguished Professor. Throughout his career, Jain has conducted pioneering research in data clustering, fingerprint recognition, and face recognition. His work has been published in leading scientific journals including Scientific American, Nature, IEEE Spectrum, and MIT Technology Review. He served as Editor-in-Chief of the IEEE Transactions on Pattern Analysis and Machine Intelligence from 1991 to 1994. Jain has also contributed to national security and policy through his service on several advisory bodies. He served as a member of the U.S. National Academies panels on Information Technology, Whither Biometrics, and Improvised Explosive Devices (IED). He has also served on the Defense Science Board, the Forensic Science Standards Board, and the AAAS Latent Fingerprint Working Group. In 2014, Jain was named Innovator of the Year at Michigan State University for transferring several technologies on face and fingerprint recognition to major players in the biometrics industry. He holds eight U.S. and Korean patents related to biometric technologies. == Research contributions == Jain's research spans pattern recognition, computer vision, machine learning, and biometric recognition. His contributions have been particularly influential in several areas: === Biometric recognition === Jain is considered one of the foremost authorities on biometric recognition systems. His research group at Michigan State University has developed algorithms and systems for fingerprint, face, and iris recognition that have been widely adopted in both academic research and commercial applications. His work on fingerprint matching algorithms has been instrumental in establishing standards for automated fingerprint identification systems (AFIS) used by law enforcement agencies worldwide. In recent years, Jain and his research team have made significant advances in child fingerprint recognition, demonstrating that digital scans of a young child's fingerprint can be correctly recognized one year later with over 99 percent accuracy for children as young as six months old. This research has important implications for child identification in developing countries, where it can be used to track immunization records and provide access to medical care. === Data clustering === Jain's survey article "Data clustering: a review" (1999), co-authored with M. N. Murty and P. J. Flynn, is one of the most highly cited papers in computer science. His 2010 paper "Data Clustering: 50 Years Beyond K-Means" provided a comprehensive overview of the evolution of clustering methods and remains an essential reference in the field. === Statistical pattern recognition === Jain's work on statistical pattern recognition, including his influential survey "Statistical pattern recognition: A review" (2000) with R. P. W. Duin and Jianchang Mao, has shaped the theoretical foundations of the field. == Citation metrics and academic impact == Jain is among the most highly cited researchers in computer science. Based on his Google Scholar profile, he had an h-index of 200 in 2020, which was the highest among computer scientists identified in a survey published by UCLA at the time. As of August 2023, his h-index on Google Scholar is 211. He has since been surpassed by Yoshua Bengio, a researcher of similar subjects (neural networks and deep learning for artificial intelligence), who had an h-index of 224 as of August 2023. Another source reported that as of December 2022, he had the highest discipline h-index (D-index) in computer science. == Honors and awards == Jain has received numerous awards and honors recognizing his contributions to computer science and engineering: === Academy memberships === Member, United States National Academy of Engineering (2016) — elected "for contributions to the engineering and practice of biometrics" Foreign Fellow, Indian National Academy of Engineering (2016) Foreign Member, Chinese Academy of Sciences (2019) Member, The World Academy of Sciences (2019) Fellow, National Academy of Inventors === Professional society fellowships === Fellow, ACM Fellow, IEEE (1988) — for contributions to image processing Fellow, AAAS Fellow, International Association for Pattern Recognition Fellow, SPIE === Major awards === BBVA Foundation Frontiers of Knowledge Award in Information and Communication Technologies (2024) IAPR King-Sun Fu Prize (2008) IEEE W. Wallace McDowell Award (2007) — the highest technical honor awarded by the IEEE Computer Society, for pioneering contributions to theory, technique, and practice of pattern recognition, computer vision, and biometric recognition systems IEEE Computer Society Technical Achievement Award (2003) IAPR Pierre Devijver Award (2002) Humboldt Research Award (2002) Guggenheim Fellowship (2001) Fulbright Fellowship (1998) IEEE ICDM Research Contribution Award (2008) === Best paper awards === IEEE Transactions on Neural Networks (1996) Pattern Recognition journal (1987, 1991, 2005) === Honorary doctorates === Universidad Autónoma de Madrid (2018) Hong Kong University of Science and Technology (2021) == Legacy and endowments == Two endowed funds have been established in Jain's honor at Michigan State University, recognizing his lasting impact on the field and the university. In 2015, a former visiting scholar from Jain's laboratory made an anonymous $400,000 gift to create the Anil K. Jain Endowed Graduate Fellowship, which supports doctoral-level research in pattern recognition, computer vision, and biometric recognition. In 2022, the Anil K. and Nandita K. Jain Endowed Professorship was established through $1 million in contributions from multiple donors, including a substantial gift from the Jain family, to support faculty recruitment and retention in the Department of Computer Science and Engineering. == Selected publications == === Books === 1988. Algorithms For Clustering Data. With Richard C. Dubes. Prentice Hall. 1993. Markov Random Fields: Theory and Applications. With Rama Chellappa eds. Academic Press. 1999. Biometrics: Personal Identification in Networked Society. With Ruud M. Bolle and Sharath Pankanti eds. Springer. 2003. Handbook of Fingerprint Recognition. (2nd edition 2009). With D. Maio, D. Maltoni, S. Prabhakar. Springer. 2005. Handbook of Face Recognition. (2nd edition 2011). With S. Z. Li ed. Springer. 2006. Handbook of Multibiometrics. With A. Ross and K. Nandakumar. Springer. 2007. Handbook of Biometrics. With P. Flynn and A. Ross eds. Springer. 2011. Introduction to Biometrics. With A. Ross and K. Nandakumar. Springer. 2015. Encyclopedia of Biometrics (Second Edition). With Stan Li. Springer. === Research articles === Cross, George R. and Anil K. Jain. "Markov random field texture models". IEEE Transactions on Pattern Analysis and Machine Intelligence (1983): 25–39. Jain, Anil K., and Farshid Farrokhnia. "Unsupervised texture segmentation using Gabor filters". Pattern Recognition 24.12 (1991): 1167–1186. Jain, Anil K., and Douglas Zongker. "Feature selection: Evaluation, application, and small sample performance". IEEE Transactions on Pattern Analysis and Machine Intelligence, 19.2 (1997): 153–158. Jain, Anil K., L. Hong, S. Pankanti, R. Bolle. "An Identity-A

  Read more →
• 

  Line integral convolution

  In scientific visualization, line integral convolution (LIC) is a method to visualize a vector field (such as fluid motion) at high spatial resolutions. The LIC technique was first proposed by Brian Cabral and Leith Casey Leedom in 1993. In LIC, discrete numerical line integration is performed along the field lines (curves) of the vector field on a uniform grid. The integral operation is a convolution of a filter kernel and an input texture, often white noise. In signal processing, this process is known as a discrete convolution. == Overview == Traditional visualizations of vector fields use small arrows or lines to represent vector direction and magnitude. This method has a low spatial resolution, which limits the density of presentable data and risks obscuring characteristic features in the data. More sophisticated methods, such as streamlines and particle tracing techniques, can be more revealing but are highly dependent on proper seed points. Texture-based methods, like LIC, avoid these problems since they depict the entire vector field at point-like (pixel) resolution. Compared to other integration-based techniques that compute field lines of the input vector field, LIC has the advantage that all structural features of the vector field are displayed, without the need to adapt the start and end points of field lines to the specific vector field. In other words, it shows the topology of the vector field. In user testing, LIC was found to be particularly good for identifying critical points. == Algorithm == === Informal description === LIC causes output values to be strongly correlated along the field lines, but uncorrelated in orthogonal directions. As a result, the field lines contrast each other and stand out visually from the background. Intuitively, the process can be understood with the following example: the flow of a vector field can be visualized by overlaying a fixed, random pattern of dark and light paint. As the flow passes by the paint, the fluid picks up some of the paint's color, averaging it with the color it has already acquired. The result is a randomly striped, smeared texture where points along the same streamline tend to have a similar color. Other physical examples include: whorl patterns of paint, oil, or foam on a river visualisation of magnetic field lines using randomly distributed iron filings fine sand being blown by strong wind === Formal mathematical description === Although the input vector field and the result image are discretized, it pays to look at it from a continuous viewpoint. Let v {\displaystyle \mathbf {v} } be the vector field given in some domain Ω {\displaystyle \Omega } . Although the input vector field is typically discretized, we regard the field v {\displaystyle \mathbf {v} } as defined in every point of Ω {\displaystyle \Omega } , i.e. we assume an interpolation. Streamlines, or more generally field lines, are tangent to the vector field in each point. They end either at the boundary of Ω {\displaystyle \Omega } or at critical points where v = 0 {\displaystyle \mathbf {v} =\mathbf {0} } . For the sake of simplicity, critical points and boundaries are ignored in the following. A field line σ {\displaystyle {\boldsymbol {\sigma }}} , parametrized by arc length s {\displaystyle s} , is defined as d σ ( s ) d s = v ( σ ( s ) ) | v ( σ ( s ) ) | . {\displaystyle {\frac {d{\boldsymbol {\sigma }}(s)}{ds}}={\frac {\mathbf {v} ({\boldsymbol {\sigma }}(s))}{|\mathbf {v} ({\boldsymbol {\sigma }}(s))|}}.} Let σ r ( s ) {\displaystyle {\boldsymbol {\sigma }}_{\mathbf {r} }(s)} be the field line that passes through the point r {\displaystyle \mathbf {r} } for s = 0 {\displaystyle s=0} . Then the image gray value at r {\displaystyle \mathbf {r} } is set to D ( r ) = ∫ − L / 2 L / 2 k ( s ) N ( σ r ( s ) ) d s {\displaystyle D(\mathbf {r} )=\int _{-L/2}^{L/2}k(s)N({\boldsymbol {\sigma }}_{\mathbf {r} }(s))ds} where k ( s ) {\displaystyle k(s)} is the convolution kernel, N ( r ) {\displaystyle N(\mathbf {r} )} is the noise image, and L {\displaystyle L} is the length of field line segment that is followed. D ( r ) {\displaystyle D(\mathbf {r} )} has to be computed for each pixel in the LIC image. If carried out naively, this is quite expensive. First, the field lines have to be computed using a numerical method for solving ordinary differential equations, like a Runge–Kutta method, and then for each pixel the convolution along a field line segment has to be calculated. The final image will normally be colored in some way. Typically, some scalar field in Ω {\displaystyle \Omega } (like the vector length) is used to determine the hue, while the grayscale LIC output determines the brightness. Different choices of convolution kernels and random noise produce different textures; for example, pink noise produces a cloudy pattern where areas of higher flow stand out as smearing, suitable for weather visualization. Further refinements in the convolution can improve the quality of the image. === Programming description === Algorithmically, LIC takes a vector field and noise texture as input, and outputs a texture. The process starts by generating in the domain of the vector field a random gray level image at the desired output resolution. Then, for every pixel in this image, the forward and backward streamline of a fixed arc length is calculated. The value assigned to the current pixel is computed by a convolution of a suitable convolution kernel with the gray levels of all the noise pixels lying on a segment of this streamline. This creates a gray level LIC image. == Versions == === Basic === Basic LIC images are grayscale images, without color and animation. While such LIC images convey the direction of the field vectors, they do not indicate orientation; for stationary fields, this can be remedied by animation. Basic LIC images do not show the length of the vectors (or the strength of the field). === Color === The length of the vectors (or the strength of the field) is usually coded in color; alternatively, animation can be used. === Animation === LIC images can be animated by using a kernel that changes over time. Samples at a constant time from the streamline would still be used, but instead of averaging all pixels in a streamline with a static kernel, a ripple-like kernel constructed from a periodic function multiplied by a Hann function acting as a window (in order to prevent artifacts) is used. The periodic function is then shifted along the period to create an animation. === Fast LIC (FLIC) === The computation can be significantly accelerated by re-using parts of already computed field lines, specializing to a box function as convolution kernel k ( s ) {\displaystyle k(s)} and avoiding redundant computations during convolution. The resulting fast LIC method can be generalized to convolution kernels that are arbitrary polynomials. === Oriented Line Integral Convolution (OLIC) === Because LIC does not encode flow orientation, it cannot distinguish between streamlines of equal direction but opposite orientation. Oriented Line Integral Convolution (OLIC) solves this issue by using a ramp-like asymmetric kernel and a low-density noise texture. The kernel asymmetrically modulates the intensity along the streamline, producing a trace that encodes orientation; the low-density of the noise texture prevents smeared traces from overlapping, aiding readability. Fast Rendering of Oriented Line Integral Convolution (FROLIC) is a variation that approximates OLIC by rendering each trace in discrete steps instead of as a continuous smear. === Unsteady Flow LIC (UFLIC) === For time-dependent vector fields (unsteady flow), a variant called Unsteady Flow LIC has been designed that maintains the coherence of the flow animation. An interactive GPU-based implementation of UFLIC has been presented. === Parallel === Since the computation of an LIC image is expensive but inherently parallel, the process has been parallelized and, with availability of GPU-based implementations, interactive on PCs. === Multidimensional === Note that the domain Ω {\displaystyle \Omega } does not have to be a 2D domain: the method is applicable to higher dimensional domains using multidimensional noise fields. However, the visualization of the higher-dimensional LIC texture is problematic; one way is to use interactive exploration with 2D slices that are manually positioned and rotated. The domain Ω {\displaystyle \Omega } does not have to be flat either; the LIC texture can be computed also for arbitrarily shaped 2D surfaces in 3D space. == Applications == This technique has been applied to a wide range of problems since it first was published in 1993, both scientific and creative, including: Representing vector fields: visualization of steady (time-independent) flows (streamlines) visual exploration of 2D autonomous dynamical systems wind mapping water flow mapping Artistic effects for image generation and stylization: pencil drawing (auto

  Read more →
• 

  RAMnets

  RAMnets is one of the oldest practical neurally inspired classification algorithms. The RAMnets is also known as a type of "n-tuple recognition method" or "weightless neural network". == Algorithm == Consider (let us say N) sets of n distinct bit locations are selected randomly. These are the n-tuples. The restriction of a pattern to an n-tuple can be regarded as an n-bit number which, together with the identity of the n-tuple, constitutes a `feature' of the pattern. The standard n-tuple recognizer operates simply as follows: A pattern is classified as belonging to the class for which it has the most features in common with at least one training pattern of that class. This is the Θ {\displaystyle \Theta } = 0 case of a more general rule whereby the class assigned to unclassified pattern u is a c r g m a x ( ∑ i = 1 N Θ ( ∑ v ∈ D c δ ( α i ( u ) , α i ( v ) ) ) ) {\displaystyle {\begin{aligned}{\underset {c}{a}}rgmax(\sum _{i=1}^{N}\Theta (\sum _{v\in D_{c}}\delta (\alpha _{i}(u),\alpha _{i}(v))))\end{aligned}}} where Dc is the set of training patterns in class c, Θ ( x ) {\displaystyle \Theta (x)} = x for 0 ≤ x ≤ θ {\displaystyle 0\leq x\leq \theta } , Θ ( x ) = θ {\displaystyle \Theta (x)=\theta } for x ≥ θ {\displaystyle x\geq \theta } , δ i , j {\displaystyle \delta _{i,j}} is the Kronecker delta( δ i , j {\displaystyle \delta _{i,j}} =1 if i=j and 0 otherwise.)and ( α i ( u ) ) {\displaystyle (\alpha _{i}(u))} is the ith feature of the pattern u: ∑ j = 0 n − 1 u η i ( j ) 2 j {\displaystyle \sum _{j=0}^{n-1}u_{\eta }i(j)2^{j}} Here uk is the kth bit of u and u η i ( j ) {\displaystyle u_{\eta }i(j)} is the jth bit location of the ith n-tuple. With C classes to distinguish, the system can be implemented as a network of NC nodes, each of which is a random access memory (RAM); hence the term RAMnet. The memory content m c i α {\displaystyle m_{ci\alpha }} at address α {\displaystyle \alpha } of the ith node allocated to class c is set to m c i α {\displaystyle m_{ci\alpha }} = Θ ( ∑ v ∈ D c δ ( α , α i ( v ) ) ) {\displaystyle \Theta (\sum _{v\in D_{c}}\delta (\alpha ,\alpha _{i}(v)))} In the usual θ {\displaystyle \theta } = 1 case, the 1-bit content of m c i α {\displaystyle m_{ci\alpha }} is set if any pattern of Dc has feature α {\displaystyle \alpha } and unset otherwise. Recognition is accomplished by summing the contents of the nodes of each class at the addresses given by the features of the unclassified pattern. That is, pattern u is assigned to class a c r g m a x ( ∑ i = 1 N m c i α ( u ) ) {\displaystyle {\begin{aligned}{\underset {c}{a}}rgmax(\sum _{i=1}^{N}m_{ci\alpha }(u))\end{aligned}}} == RAM-discriminators and WiSARD == The RAMnets formed the basis of a commercial product known as WiSARD (Wilkie, Stonham and Aleksander Recognition Device) was the first artificial neural network machine to be patented. A RAM-discriminator consists of a set of X one-bit word RAMs with n inputs and a summing device (Σ). Any such RAM-discriminator can receive a binary pattern of X⋅n bits as input. The RAM input lines are connected to the input pattern by means of a biunivocal pseudo-random mapping. The summing device enables this network of RAMs to exhibit – just like other ANN models based on synaptic weights – generalization and noise tolerance. In order to train the discriminator one has to set all RAM memory locations to 0 and choose a training set formed by binary patterns of X⋅n bits. For each training pattern, a 1 is stored in the memory location of each RAM addressed by this input pattern. Once the training of patterns is completed, RAM memory contents will be set to a certain number of 0's and 1's. The information stored by the RAM during the training phase is used to deal with previous unseen patterns. When one of these is given as input, the RAM memory contents addressed by the input pattern are read and summed by Σ. The number r thus obtained, which is called the discriminator response, is equal to the number of RAMs that output 1. r reaches the maximum X if the input belongs to the training set. r is equal to 0 if no n-bit component of the input pattern appears in the training set (not a single RAM outputs 1). Intermediate values of r express a kind of “similarity measure” of the input pattern with respect to the patterns in the training set. A system formed by various RAM-discriminators is called WiSARD. Each RAM-discriminator is trained on a particular class of patterns, and classification by the multi-discriminator system is performed in the following way. When a pattern is given as input, each RAM-discriminator gives a response to that input. The various responses are evaluated by an algorithm which compares them and computes the relative confidence c of the highest response (e.g., the difference d between the highest response and the second highest response, divided by the highest response). A schematic representation of a RAM-discriminator and a 10 RAM-discriminator WiSARD is shown in Figure 1.

  Read more →
• 

  Trie

  In computer science, a trie (, ), also known as a digital tree or prefix tree, is a specialized search tree data structure used to store and retrieve strings from a dictionary or set. Unlike a binary search tree, nodes in a trie do not store their associated key. Instead, each node's position within the trie determines its associated key, with the connections between nodes defined by individual characters rather than the entire key. Tries are particularly effective for tasks such as autocomplete, spell checking, and IP routing, offering advantages over hash tables due to their prefix-based organization and lack of hash collisions. Every child node shares a common prefix with its parent node, and the root node represents the empty string. While basic trie implementations can be memory-intensive, various optimization techniques such as compression and bitwise representations have been developed to improve their efficiency. A notable optimization is the radix tree, which provides more efficient prefix-based storage. While tries store character strings, they can be adapted to work with any ordered sequence of elements, such as permutations of digits or shapes. A notable variant is the bitwise trie, which uses individual bits from fixed-length binary data (such as integers or memory addresses) as keys. == History, etymology, and pronunciation == The idea of a trie for representing a set of strings was first abstractly described by Axel Thue in 1912. Tries were first described in a computer context by René de la Briandais in 1959. The idea was independently described in 1960 by Edward Fredkin, who coined the term trie, pronouncing it (as "tree"), after the middle syllable of retrieval. However, other authors pronounce it (as "try"), in an attempt to distinguish it verbally from "tree". == Overview == Tries are a form of string-indexed look-up data structure, which is used to store a dictionary list of words that can be searched on in a manner that allows for efficient generation of completion lists. A prefix trie is an ordered tree data structure used in the representation of a set of strings over a finite alphabet set, which allows efficient storage of words with common prefixes. Tries can be efficacious on string-searching algorithms such as predictive text, approximate string matching, and spell checking in comparison to binary search trees. A trie can be seen as a tree-shaped deterministic finite automaton. == Operations == Tries support various operations: insertion, deletion, and lookup of a string key. Tries are composed of nodes that contain links, which either point to other suffix child nodes or null. As for every tree, each node except the root is pointed to by only one other node, called its parent. Each node contains as many links as the number of characters in the applicable alphabet (although tries tend to have a substantial number of null links). In some cases, the alphabet used is simply that of the character encoding—resulting in, for example, a size of 128 in the case of ASCII. The null links within the children of a node emphasize the following characteristics: Characters and string keys are implicitly stored in the trie, and include a character sentinel value indicating string termination. Each node contains one possible link to a prefix of strong keys of the set. A basic structure type of nodes in the trie is as follows: Node {\displaystyle {\text{Node}}} may contain an optional Value {\displaystyle {\text{Value}}} , which is associated with the key that corresponds to the node. === Searching === Searching for a value in a trie is guided by the characters in the search string key, as each node in the trie contains a corresponding link to each possible character in the given string. Thus, following the string within the trie yields the associated value for the given string key. A null link during the search indicates the inexistence of the key. The following pseudocode implements the search procedure for a given string key in a rooted trie x. In the above pseudocode, x and key correspond to the pointer of the trie's root node and the string key, respectively. The search operation takes O ( m ) {\displaystyle O(m)} time, where m {\displaystyle m} is the size of the string parameter key. In a balanced binary search tree, on the other hand, it takes O ( m log ⁡ n ) {\displaystyle O(m\log n)} time, in the worst case, since key needs to be compared with O ( log ⁡ n ) {\displaystyle O(\log n)} other keys and each comparison takes O ( m ) {\displaystyle O(m)} time, in the worst case. The trie occupies less space, in comparison with a binary search tree, in the case of a large number of short strings, since nodes share common initial string subsequences and store the keys implicitly. === Insertion === Insertion into a trie is guided by using the character sets as indexes to the children array until the last character of the string key is reached. Each node in the trie corresponds to one call of the radix sorting routine, as the trie structure reflects the execution pattern of the top-down radix sort. If null links are encountered before reaching the last character of the string key, new nodes are created. The input value is assigned to the value of the last node traversed, which is the node that corresponds to the key. === Deletion === Deletion of a key–value pair from a trie involves finding the node corresponding to the key, setting its value to null, and recursively removing nodes that have no children. The procedure begins by examining key; an empty string indicates arrival at the node corresponding to the (original) key, in which case its value is set to null. If the node, then, has null value and no children, it is removed from the trie by returning null; otherwise, the node is kept by returning the node itself. == Replacing other data structures == === Replacement for hash tables === A trie can be used to replace a hash table, over which it has the following advantages: Searching for a node with an associated key of size m {\displaystyle m} has the complexity of O ( m ) {\displaystyle O(m)} , whereas an imperfect hash function may have numerous colliding keys, and the worst-case lookup speed of such a table would be O ( N ) {\displaystyle O(N)} , where N {\displaystyle N} denotes the total number of nodes within the table. Tries do not need a hash function for the operation, unlike a hash table; there are also no collisions of different keys in a trie. Within a trie, keys can be efficiently sorted lexicographically. However, tries are less efficient than a hash table when the data is directly accessed on a secondary storage device such as a hard disk drive that has higher random access time than the main memory. == Implementation strategies == Tries can be represented in several ways, corresponding to different trade-offs between memory use and speed of the operations. Using a vector of pointers for representing a trie consumes enormous space; however, memory space can be reduced at the expense of running time if a singly linked list is used for each node vector, as most entries of the vector contains nil {\displaystyle {\text{nil}}} . Techniques such as alphabet reduction may reduce the large space requirements by reinterpreting the original string as a longer string over a smaller alphabet. For example, a string of n bytes can alternatively be regarded as a string of 2n four-bit units. This can reduce memory usage by a factor of eight; but lookups need to visit twice as many nodes in the worst case. Another technique includes storing a vector of 256 ASCII pointers as a bitmap of 256 bits representing ASCII alphabet, which reduces the size of individual nodes dramatically. === Bitwise tries === Bitwise tries are used to address the enormous space requirement for the trie nodes in a naive simple pointer vector implementations. Each character in the string key set is represented via individual bits, which are used to traverse the trie over a string key. The implementations for these types of trie use vectorized CPU instructions to find the first set bit in a fixed-length key input (e.g. GCC's __builtin_clz() intrinsic function). Accordingly, the set bit is used to index the first item, or child node, in the 32- or 64-entry based bitwise tree. Search then proceeds by testing each subsequent bit in the key. This procedure is also cache-local and highly parallelizable due to register independency, and thus performant on out-of-order execution CPUs. === Compressed tries === Radix tree, also known as a compressed trie, is a space-optimized variant of a trie in which any node with only one child gets merged with its parent; elimination of branches of the nodes with a single child results in better metrics in both space and time. This works best when the trie remains static and set of keys stored are very sparse within their representation space. One more approach for static tries is to "pack" the trie by storing disjoint

  Read more →