In dual decomposition a problem is broken into smaller subproblems and a solution to the relaxed problem is found. This method can be employed for MRF optimization. Dual decomposition is applied to markov logic programs as an inference technique. == Background == Discrete MRF Optimization (inference) is very important in Machine Learning and Computer vision, which is realized on CUDA graphical processing units. Consider a graph G = ( V , E ) {\displaystyle G=(V,E)} with nodes V {\displaystyle V} and Edges E {\displaystyle E} . The goal is to assign a label l p {\displaystyle l_{p}} to each p ∈ V {\displaystyle p\in V} so that the MRF Energy is minimized: (1) min Σ p ∈ V θ p ( l p ) + Σ p q ∈ ε θ p q ( l p ) ( l q ) {\displaystyle \min \Sigma _{p\in V}\theta _{p}(l_{p})+\Sigma _{pq\in \varepsilon }\theta _{pq}(l_{p})(l_{q})} Major MRF Optimization methods are based on Graph cuts or Message passing. They rely on the following integer linear programming formulation (2) min x E ( θ , x ) = θ . x = ∑ p ∈ V θ p . x p + ∑ p q ∈ ε θ p q . x p q {\displaystyle \min _{x}E(\theta ,x)=\theta .x=\sum _{p\in V}\theta _{p}.x_{p}+\sum _{pq\in \varepsilon }\theta _{pq}.x_{pq}} In many applications, the MRF-variables are {0,1}-variables that satisfy: x p ( l ) = 1 {\displaystyle x_{p}(l)=1} ⇔ {\displaystyle \Leftrightarrow } label l {\displaystyle l} is assigned to p {\displaystyle p} , while x p q ( l , l ′ ) = 1 {\displaystyle x_{pq}(l,l^{\prime })=1} , labels l , l ′ {\displaystyle l,l^{\prime }} are assigned to p , q {\displaystyle p,q} . == Dual Decomposition == The main idea behind decomposition is surprisingly simple: decompose your original complex problem into smaller solvable subproblems, extract a solution by cleverly combining the solutions from these subproblems. A sample problem to decompose: min x Σ i f i ( x ) {\displaystyle \min _{x}\Sigma _{i}f^{i}(x)} where x ∈ C {\displaystyle x\in C} In this problem, separately minimizing every single f i ( x ) {\displaystyle f^{i}(x)} over x {\displaystyle x} is easy; but minimizing their sum is a complex problem. So the problem needs to get decomposed using auxiliary variables { x i } {\displaystyle \{x^{i}\}} and the problem will be as follows: min { x i } , x Σ i f i ( x i ) {\displaystyle \min _{\{x^{i}\},x}\Sigma _{i}f^{i}(x^{i})} where x i ∈ C , x i = x {\displaystyle x^{i}\in C,x^{i}=x} Now we can relax the constraints by multipliers { λ i } {\displaystyle \{\lambda ^{i}\}} which gives us the following Lagrangian dual function: g ( { λ i } ) = min { x i ∈ C } , x Σ i f i ( x i ) + Σ i λ i . ( x i − x ) = min { x i ∈ C } , x Σ i [ f i ( x i ) + λ i . x i ] − ( Σ i λ i ) x {\displaystyle g(\{\lambda ^{i}\})=\min _{\{x^{i}\in C\},x}\Sigma _{i}f^{i}(x^{i})+\Sigma _{i}\lambda ^{i}.(x^{i}-x)=\min _{\{x^{i}\in C\},x}\Sigma _{i}[f^{i}(x^{i})+\lambda ^{i}.x^{i}]-(\Sigma _{i}\lambda ^{i})x} Now we eliminate x {\displaystyle x} from the dual function by minimizing over x {\displaystyle x} and dual function becomes: g ( { λ i } ) = min { x i ∈ C } Σ i [ f i ( x i ) + λ i . x i ] {\displaystyle g(\{\lambda ^{i}\})=\min _{\{x^{i}\in C\}}\Sigma _{i}[f^{i}(x^{i})+\lambda ^{i}.x^{i}]} We can set up a Lagrangian dual problem: (3) max { λ i } ∈ Λ g ( λ i ) = Σ i g i ( x i ) , {\displaystyle \max _{\{\lambda ^{i}\}\in \Lambda }g({\lambda ^{i}})=\Sigma _{i}g^{i}(x^{i}),} The Master problem (4) g i ( x i ) = m i n x i f i ( x i ) + λ i . x i {\displaystyle g^{i}(x^{i})=min_{x^{i}}f^{i}(x^{i})+\lambda ^{i}.x^{i}} where x i ∈ C {\displaystyle x^{i}\in C} The Slave problems == MRF optimization via Dual Decomposition == The original MRF optimization problem is NP-hard and we need to transform it into something easier. τ {\displaystyle \tau } is a set of sub-trees of graph G {\displaystyle G} where its trees cover all nodes and edges of the main graph. And MRFs defined for every tree T {\displaystyle T} in τ {\displaystyle \tau } will be smaller. The vector of MRF parameters is θ T {\displaystyle \theta ^{T}} and the vector of MRF variables is x T {\displaystyle x^{T}} , these two are just smaller in comparison with original MRF vectors θ , x {\displaystyle \theta ,x} . For all vectors θ T {\displaystyle \theta ^{T}} we'll have the following: (5) ∑ T ∈ τ ( p ) θ p T = θ p , ∑ T ∈ τ ( p q ) θ p q T = θ p q . {\displaystyle \sum _{T\in \tau (p)}\theta _{p}^{T}=\theta _{p},\sum _{T\in \tau (pq)}\theta _{pq}^{T}=\theta _{pq}.} Where τ ( p ) {\displaystyle \tau (p)} and τ ( p q ) {\displaystyle \tau (pq)} denote all trees of τ {\displaystyle \tau } than contain node p {\displaystyle p} and edge p q {\displaystyle pq} respectively. We simply can write: (6) E ( θ , x ) = ∑ T ∈ τ E ( θ T , x T ) {\displaystyle E(\theta ,x)=\sum _{T\in \tau }E(\theta ^{T},x^{T})} And our constraints will be: (7) x T ∈ χ T , x T = x | T , ∀ T ∈ τ {\displaystyle x^{T}\in \chi ^{T},x^{T}=x_{|T},\forall T\in \tau } Our original MRF problem will become: (8) min { x T } , x Σ T ∈ τ E ( θ T , x T ) {\displaystyle \min _{\{x^{T}\},x}\Sigma _{T\in \tau }E(\theta ^{T},x^{T})} where x T ∈ χ T , ∀ T ∈ τ {\displaystyle x^{T}\in \chi ^{T},\forall T\in \tau } and x T ∈ x | T , ∀ T ∈ τ {\displaystyle x^{T}\in x_{|T},\forall T\in \tau } And we'll have the dual problem we were seeking: (9) max { λ T } ∈ Λ g ( { λ T } ) = ∑ T ∈ τ g T ( λ T ) , {\displaystyle \max _{\{\lambda ^{T}\}\in \Lambda }g(\{\lambda ^{T}\})=\sum _{T\in \tau }g^{T}(\lambda ^{T}),} The Master problem where each function g T ( . ) {\displaystyle g^{T}(.)} is defined as: (10) g T ( λ T ) = min x T E ( θ T + λ T , x T ) {\displaystyle g^{T}(\lambda ^{T})=\min _{x^{T}}E(\theta ^{T}+\lambda ^{T},x^{T})} where x T ∈ χ T {\displaystyle x^{T}\in \chi ^{T}} The Slave problems == Theoretical Properties == Theorem 1. Lagrangian relaxation (9) is equivalent to the LP relaxation of (2). min { x T } , x { E ( x , θ ) | x p T = s p , x T ∈ CONVEXHULL ( χ T ) } {\displaystyle \min _{\{x^{T}\},x}\{E(x,\theta )|x_{p}^{T}=s_{p},x^{T}\in {\text{CONVEXHULL}}(\chi ^{T})\}} Theorem 2. If the sequence of multipliers { α t } {\displaystyle \{\alpha _{t}\}} satisfies α t ≥ 0 , lim t → ∞ α t = 0 , ∑ t = 0 ∞ α t = ∞ {\displaystyle \alpha _{t}\geq 0,\lim _{t\to \infty }\alpha _{t}=0,\sum _{t=0}^{\infty }\alpha _{t}=\infty } then the algorithm converges to the optimal solution of (9). Theorem 3. The distance of the current solution { θ T } {\displaystyle \{\theta ^{T}\}} to the optimal solution { θ ¯ T } {\displaystyle \{{\bar {\theta }}^{T}\}} , which decreases at every iteration. Theorem 4. Any solution obtained by the method satisfies the WTA (weak tree agreement) condition. Theorem 5. For binary MRFs with sub-modular energies, the method computes a globally optimal solution.
Hedgeable
Hedgeable, Inc. was a U.S. based financial services company and digital wealth management platform headquartered in New York City. Hedgeable was known for not following set allocations, and instead actively managing accounts in response to market movements. On August 9, 2018, Hedgeable closed its doors to new investors, with existing investors required to transfer out of the company. The company claimed that it was not shutting down but simply removing its SEC registration. == History == Hedgeable was founded in 2009 by twin brothers Michael and Matthew Kane, who previously worked at high-net worth investment managers such as Bridgewater Associates and Spruce Private Investors. Both Michael and Matthew graduated from Penn State University with degrees in finance. Hedgeable is a Registered Investment Advisor with the U.S. Securities and Exchange Commission. The company has received funding from SixThirty and Route 66 Ventures as well as various other angel investors. On August 9, 2018, Hedgeable closed its doors to new investors. == Investing Strategies == Hedgeable did not follow a buy-and-hold approach, but instead actively manages accounts in response to market movements focusing on downside protection in bear markets. Their strategy was different from other robo-advisors, which use Modern Portfolio Theory. Hedgeable offered investment options including Exchange Traded Funds (ETFs) to individual stocks, master limited partnerships, private equity and bitcoin. Mutual funds were not used in portfolios. Although the firm's focus was to provide a direct-to-consumer service, Hedgeable's investment strategies were available to financial advisors and institutions as well through a variety of platforms. == Product Features == When it was open to external clients, Hedgeable aimed to gamify their personal finance experience. Clients could open a new account or transfer an existing account. Hedgeable accepted retirement accounts, taxable accounts, business accounts and various other account types. Hedgeable offered the following features: Downside protection Account aggregation Alternative investments Alpha rewards API Mobile app It was awarded 4/5 for client transparency by Paladin Research. Hedgeable was the winner of the Finovate Fall 2015 Best of Show Award and the GREAT 2015 Tech Award (FinTech Category). In 2016, Hedgeable launched its first iOS mobile app in order to expand their product offerings.
Factorization of polynomials over finite fields
In mathematics and computer algebra the factorization of a polynomial consists of decomposing it into a product of irreducible factors. This decomposition is theoretically possible and is unique for polynomials with coefficients in any field, but rather strong restrictions on the field of the coefficients are needed to allow the computation of the factorization by means of an algorithm. In practice, algorithms have been designed only for polynomials with coefficients in a finite field, in the field of rationals or in a finitely generated field extension of one of them. All factorization algorithms, including the case of multivariate polynomials over the rational numbers, reduce the problem to this case; see polynomial factorization. It is also used for various applications of finite fields, such as coding theory (cyclic redundancy codes and BCH codes), cryptography (public key cryptography by the means of elliptic curves), and computational number theory. As the reduction of the factorization of multivariate polynomials to that of univariate polynomials does not have any specificity in the case of coefficients in a finite field, only polynomials with one variable are considered in this article. == Background == === Finite field === The theory of finite fields, whose origins can be traced back to the works of Gauss and Galois, has played a part in various branches of mathematics. Due to the applicability of the concept in other topics of mathematics and sciences like computer science there has been a resurgence of interest in finite fields and this is partly due to important applications in coding theory and cryptography. Applications of finite fields introduce some of these developments in cryptography, computer algebra and coding theory. A finite field or Galois field is a field with a finite order (number of elements). The order of a finite field is always a prime or a power of prime. For each prime power q = pr, there exists exactly one finite field with q elements, up to isomorphism. This field is denoted GF(q) or Fq. If p is prime, GF(p) is the prime field of order p; it is the field of residue classes modulo p, and its p elements are denoted 0, 1, ..., p−1. Thus a = b in GF(p) means the same as a ≡ b (mod p). === Irreducible polynomials === Let F be a finite field. As for general fields, a non-constant polynomial f in F[x] is said to be irreducible over F if it is not the product of two polynomials of positive degree. A polynomial of positive degree that is not irreducible over F is called reducible over F. Irreducible polynomials allow us to construct the finite fields of non-prime order. In fact, for a prime power q, let Fq be the finite field with q elements, unique up to isomorphism. A polynomial f of degree n greater than one, which is irreducible over Fq, defines a field extension of degree n which is isomorphic to the field with qn elements: the elements of this extension are the polynomials of degree lower than n; addition, subtraction and multiplication by an element of Fq are those of the polynomials; the product of two elements is the remainder of the division by f of their product as polynomials; the inverse of an element may be computed by the extended GCD algorithm (see Arithmetic of algebraic extensions). It follows that, to compute in a finite field of non prime order, one needs to generate an irreducible polynomial. For this, the common method is to take a polynomial at random and test it for irreducibility. For sake of efficiency of the multiplication in the field, it is usual to search for polynomials of the shape xn + ax + b. Irreducible polynomials over finite fields are also useful for pseudorandom number generators using feedback shift registers and discrete logarithm over F2n. The number of irreducible monic polynomials of degree n over Fq is the number of aperiodic necklaces, given by Moreau's necklace-counting function Mq(n). The closely related necklace function Nq(n) counts monic polynomials of degree n which are primary (a power of an irreducible); or alternatively irreducible polynomials of all degrees d which divide n. === Example === The polynomial P = x4 + 1 is irreducible over Q but not over any finite field. On any field extension of F2, P = (x + 1)4. On every other finite field, at least one of −1, 2 and −2 is a square, because the product of two non-squares is a square and so we have If − 1 = a 2 , {\displaystyle -1=a^{2},} then P = ( x 2 + a ) ( x 2 − a ) . {\displaystyle P=(x^{2}+a)(x^{2}-a).} If 2 = b 2 , {\displaystyle 2=b^{2},} then P = ( x 2 + b x + 1 ) ( x 2 − b x + 1 ) . {\displaystyle P=(x^{2}+bx+1)(x^{2}-bx+1).} If − 2 = c 2 , {\displaystyle -2=c^{2},} then P = ( x 2 + c x − 1 ) ( x 2 − c x − 1 ) . {\displaystyle P=(x^{2}+cx-1)(x^{2}-cx-1).} === Complexity === Polynomial factoring algorithms use basic polynomial operations such as products, divisions, gcd, powers of one polynomial modulo another, etc. A multiplication of two polynomials of degree at most n can be done in O(n2) operations in Fq using "classical" arithmetic, or in O(nlog(n) log(log(n)) ) operations in Fq using "fast" arithmetic. A Euclidean division (division with remainder) can be performed within the same time bounds. The cost of a polynomial greatest common divisor between two polynomials of degree at most n can be taken as O(n2) operations in Fq using classical methods, or as O(nlog2(n) log(log(n)) ) operations in Fq using fast methods. For polynomials h, g of degree at most n, the exponentiation hq mod g can be done with O(log(q)) polynomial products, using exponentiation by squaring method, that is O(n2log(q)) operations in Fq using classical methods, or O(nlog(q)log(n) log(log(n))) operations in Fq using fast methods. In the algorithms that follow, the complexities are expressed in terms of number of arithmetic operations in Fq, using classical algorithms for the arithmetic of polynomials. == Factoring algorithms == Many algorithms for factoring polynomials over finite fields include the following three stages: Square-free factorization Distinct-degree factorization Equal-degree factorization An important exception is Berlekamp's algorithm, which combines stages 2 and 3. === Berlekamp's algorithm === Berlekamp's algorithm is historically important as being the first factorization algorithm which works well in practice. However, it contains a loop on the elements of the ground field, which implies that it is practicable only over small finite fields. For a fixed ground field, its time complexity is polynomial, but, for general ground fields, the complexity is exponential in the size of the ground field. === Square-free factorization === The algorithm determines a square-free factorization for polynomials whose coefficients come from the finite field Fq of order q = pm with p a prime. This algorithm firstly determines the derivative and then computes the gcd of the polynomial and its derivative. If it is not one then the gcd is again divided into the original polynomial, provided that the derivative is not zero (a case that exists for non-constant polynomials defined over finite fields). This algorithm uses the fact that, if the derivative of a polynomial is zero, then it is a polynomial in xp, which is, if the coefficients belong to Fp, the pth power of the polynomial obtained by substituting x by x1/p. If the coefficients do not belong to Fp, the pth root of a polynomial with zero derivative is obtained by the same substitution on x, completed by applying the inverse of the Frobenius automorphism to the coefficients. This algorithm works also over a field of characteristic zero, with the only difference that it never enters in the blocks of instructions where pth roots are computed. However, in this case, Yun's algorithm is much more efficient because it computes the greatest common divisors of polynomials of lower degrees. A consequence is that, when factoring a polynomial over the integers, the algorithm which follows is not used: one first computes the square-free factorization over the integers, and to factor the resulting polynomials, one chooses a p such that they remain square-free modulo p. Algorithm: SFF (Square-Free Factorization) Input: A monic polynomial f in Fq[x] where q = pm Output: Square-free factorization of f R ← 1 # Make w be the product (without multiplicity) of all factors of f that have # multiplicity not divisible by p c ← gcd(f, f′) w ← f/c # Step 1: Identify all factors in w i ← 1 while w ≠ 1 do y ← gcd(w, c) fac ← w / y R ← R · faci w ← y; c ← c / y; i ← i + 1 end while # c is now the product (with multiplicity) of the remaining factors of f # Step 2: Identify all remaining factors using recursion # Note that these are the factors of f that have multiplicity divisible by p if c ≠ 1 then c ← c1/p R ← R·SFF(c)p end if Output(R) The idea is to identify the product of all irreducible factors of f with the same multiplicity. This is done in two steps. The first step uses the formal d
Story (social media)
In social media, a story is a function in which the user tells a narrative or provides status messages and information in the form of short, time-limited clips in an automatically running sequence. == Definition == A story is a short sequence of images, videos, or other social media content, which can be accompanied by backgrounds, music, text, stickers, animations, filters or emojis. Social media platforms typically advance through the sequence automatically when presenting a story to a viewer. Although the sequential nature of stories can be used to tell a narrative, the pieces of a story can also be unrelated. Social media platforms that offer stories will typically have a primary story for each user which consists of everything the user posted to their story over a certain period of time, usually the most recent 24 hours. Most stories cannot be changed afterwards and are only available for a short time. Stories are almost exclusively created on a mobile device such as a smartphone or tablet computer and are usually displayed vertically. == History == In October 2013, Snapchat first introduced the story function as a series of Snaps that can together tell a narrative through a chronological order, with each Snap being viewable by all of the poster's friends and deleted after 24 hours. Stories soon surpassed private Snaps to become Snapchat's most-viewed type of post. After 2015, Snapchat introduced a feature allowing users to post private stories viewable by a chosen subset of their friends. Later other apps would copy this feature. In August 2016, Instagram introduced a stories function that deletes the content after 24 hours. Various commenters have accused the site of copying Snapchat. In February 2017, the instant messenger WhatsApp introduced the Now Status stories function in beta, which was later renamed Status. In March 2017, a story function was introduced in Facebook Messenger. In February 2018, Google launched AMP Stories, bringing a story-style format to certain Google search results on mobile devices. In August 2018, YouTube introduced a stories function that initially was limited to pictures, but was later expanded to support short video clips. The feature was shut down in June 2023. In August 2018, the GIF website Giphy introduced a story function. In March 2022, TikTok added a story feature which allowed users to create 15 second long videos that delete after 24 hours. In June 2023, Telegram CEO Pavel Durov announced stories for Telegram would be released in July 2023. In July 2023, the feature was released for premium users, and in August 2023 it was rolled out for all users. == User motivations == In 2022, a study performed by Jia-Dai (Evelyn) Lu and Jhih-Syuan (Elaine) Lin examined the various motivations for updating stories on Instagram. The researchers found a new configuration of motivations for using Instagram Stories: exploration, self-enhancement, perceived functionality, entertainment, social sharing, relationship building, novelty, and surveillance. The findings also highlighted that contribution and creation activities are likely to result in positive emotions, while creation alone predicts negative emotions while updating stories on Instagram. == Usage statistics == In 2019, around 1.5 billion people worldwide every day on average used the stories function in a social network or messenger. Younger people in particular use this function. More than 20% of people aged 18 to 24 use Instagram stories, while it is just under 2% of those over 55. In a Facebook survey of 18,000 participants from 12 countries, 68% said they used the stories function at least once a month. Stories in the areas of fashion and tourism are particularly popular. The website Fanpage Karma analyzed several Instagram accounts and determined the average reach of posts and stories per follower, concluding that posts have a higher reach than stories, which often have less than half the reach.
Data thinking
Data Thinking is a framework that integrates data science with the design process. It combines computational thinking, statistical thinking, and domain-specific knowledge to guide the development of data-driven solutions in product development. The framework is used to explore, design, develop, and validate solutions, with a focus on user experience and data analytics, including data collection and interpretation The framework aims to apply data literacy and inform decision-making through data-driven insights. == Major components == According to "Computational thinking in the era of data science": Data thinking involves understanding that solutions require both data-driven and domain-knowledge-driven rules. Data thinking evaluates whether data accurately represents real-life scenarios and improves data collection where necessary. The framework highlights the importance of preserving domain-specific meaning during data analysis. Data thinking incorporates statistical and logical analysis to identify patterns and irregularities. Data thinking involves testing solutions in real-life contexts and iteratively improving models based on new data. The process requires evaluating problems from multiple abstraction levels and understanding the potential for biases in generalizations. == Major phases == === Strategic context and risk analysis === Analyzing the broader digital strategy and assessing risks and opportunities is a common step before beginning a project. Techniques like coolhunting, trend analysis, and scenario planning can be used to assist with this. === Ideation and exploration === In this phase, focus areas are identified, and use cases are developed by integrating organizational goals, user needs, and data requirements. Design thinking methods, such as personas and customer journey mapping, are applied. === Prototyping === A proof of concept is created to test feasibility and refine solutions through iterative evaluation to optimize for effective performance. === Implementation and monitoring === Solutions are tested and monitored for performance and continual improvement. == Implementing Data Thinking == The following resources explain more about data thinking and its applications: "Data Thinking: Framework for data-based solutions" by StackFuel "What is Data Thinking? A modern approach to designing a data strategy" by Mantel Group "Data Science Thinking" by SpringerLink These sources provide detailed insights into the methodology, phases, and benefits of adopting Data Thinking in organizational processes.
Quantum natural language processing
Quantum natural language processing (QNLP) is the application of quantum computing to natural language processing (NLP). It computes word embeddings as parameterised quantum circuits that can solve NLP tasks faster than any classical computer. It is inspired by categorical quantum mechanics and the DisCoCat framework, making use of string diagrams to translate from grammatical structure to quantum processes. == Theory == The first quantum algorithm for natural language processing used the DisCoCat framework and Grover's algorithm to show a quadratic quantum speedup for a text classification task. It was later shown that quantum language processing is BQP-Complete, i.e. quantum language models are more expressive than their classical counterpart, unless quantum mechanics can be efficiently simulated by classical computers. These two theoretical results assume fault-tolerant quantum computation and a QRAM, i.e. an efficient way to load classical data on a quantum computer. Thus, they are not applicable to the noisy intermediate-scale quantum (NISQ) computers available today. == Experiments == The algorithm of Zeng and Coecke was adapted to the constraints of NISQ computers and implemented on IBM quantum computers to solve binary classification tasks. Instead of loading classical word vectors onto a quantum memory, the word vectors are computed directly as the parameters of quantum circuits. These parameters are optimised using methods from quantum machine learning to solve data-driven tasks such as question answering, machine translation and even algorithmic music composition.
Contrast set learning
Contrast set learning is a form of association rule learning that seeks to identify meaningful differences between separate groups by reverse-engineering the key predictors that identify for each particular group. For example, given a set of attributes for a pool of students (labeled by degree type), a contrast set learner would identify the contrasting features between students seeking bachelor's degrees and those working toward PhD degrees. == Overview == A common practice in data mining is to classify, to look at the attributes of an object or situation and make a guess at what category the observed item belongs to. As new evidence is examined (typically by feeding a training set to a learning algorithm), these guesses are refined and improved. Contrast set learning works in the opposite direction. While classifiers read a collection of data and collect information that is used to place new data into a series of discrete categories, contrast set learning takes the category that an item belongs to and attempts to reverse engineer the statistical evidence that identifies an item as a member of a class. That is, contrast set learners seek rules associating attribute values with changes to the class distribution. They seek to identify the key predictors that contrast one classification from another. For example, an aerospace engineer might record data on test launches of a new rocket. Measurements would be taken at regular intervals throughout the launch, noting factors such as the trajectory of the rocket, operating temperatures, external pressures, and so on. If the rocket launch fails after a number of successful tests, the engineer could use contrast set learning to distinguish between the successful and failed tests. A contrast set learner will produce a set of association rules that, when applied, will indicate the key predictors of each failed tests versus the successful ones (the temperature was too high, the wind pressure was too high, etc.). Contrast set learning is a form of association rule learning. Association rule learners typically offer rules linking attributes commonly occurring together in a training set (for instance, people who are enrolled in four-year programs and take a full course load tend to also live near campus). Instead of finding rules that describe the current situation, contrast set learners seek rules that differ meaningfully in their distribution across groups (and thus, can be used as predictors for those groups). For example, a contrast set learner could ask, “What are the key identifiers of a person with a bachelor's degree or a person with a PhD, and how do people with PhD's and bachelor’s degrees differ?” Standard classifier algorithms, such as C4.5, have no concept of class importance (that is, they do not know if a class is "good" or "bad"). Such learners cannot bias or filter their predictions towards certain desired classes. As the goal of contrast set learning is to discover meaningful differences between groups, it is useful to be able to target the learned rules towards certain classifications. Several contrast set learners, such as MINWAL or the family of TAR algorithms, assign weights to each class in order to focus the learned theories toward outcomes that are of interest to a particular audience. Thus, contrast set learning can be thought of as a form of weighted class learning. === Example: Supermarket Purchases === The differences between standard classification, association rule learning, and contrast set learning can be illustrated with a simple supermarket metaphor. In the following small dataset, each row is a supermarket transaction and each "1" indicates that the item was purchased (a "0" indicates that the item was not purchased): Given this data, Association rule learning may discover that customers that buy onions and potatoes together are likely to also purchase hamburger meat. Classification may discover that customers that bought onions, potatoes, and hamburger meats were purchasing items for a cookout. Contrast set learning may discover that the major difference between customers shopping for a cookout and those shopping for an anniversary dinner are that customers acquiring items for a cookout purchase onions, potatoes, and hamburger meat (and do not purchase foie gras or champagne). == Treatment learning == Treatment learning is a form of weighted contrast-set learning that takes a single desirable group and contrasts it against the remaining undesirable groups (the level of desirability is represented by weighted classes). The resulting "treatment" suggests a set of rules that, when applied, will lead to the desired outcome. Treatment learning differs from standard contrast set learning through the following constraints: Rather than seeking the differences between all groups, treatment learning specifies a particular group to focus on, applies a weight to this desired grouping, and lumps the remaining groups into one "undesired" category. Treatment learning has a stated focus on minimal theories. In practice, treatment are limited to a maximum of four constraints (i.e., rather than stating all of the reasons that a rocket differs from a skateboard, a treatment learner will state one to four major differences that predict for rockets at a high level of statistical significance). This focus on simplicity is an important goal for treatment learners. Treatment learning seeks the smallest change that has the greatest impact on the class distribution. Conceptually, treatment learners explore all possible subsets of the range of values for all attributes. Such a search is often infeasible in practice, so treatment learning often focuses instead on quickly pruning and ignoring attribute ranges that, when applied, lead to a class distribution where the desired class is in the minority. === Example: Boston housing data === The following example demonstrates the output of the treatment learner TAR3 on a dataset of housing data from the city of Boston (a nontrivial public dataset with over 500 examples). In this dataset, a number of factors are collected for each house, and each house is classified according to its quality (low, medium-low, medium-high, and high). The desired class is set to "high", and all other classes are lumped together as undesirable. The output of the treatment learner is as follows: Baseline class distribution: low: 29% medlow: 29% medhigh: 21% high: 21% Suggested Treatment: [PTRATIO=[12.6..16), RM=[6.7..9.78)] New class distribution: low: 0% medlow: 0% medhigh: 3% high: 97% With no applied treatments (rules), the desired class represents only 21% of the class distribution. However, if one filters the data set for houses with 6.7 to 9.78 rooms and a neighborhood parent-teacher ratio of 12.6 to 16, then 97% of the remaining examples fall into the desired class (high-quality houses). == Algorithms == There are a number of algorithms that perform contrast set learning. The following subsections describe two examples. === STUCCO === The STUCCO contrast set learner treats the task of learning from contrast sets as a tree search problem where the root node of the tree is an empty contrast set. Children are added by specializing the set with additional items picked through a canonical ordering of attributes (to avoid visiting the same nodes twice). Children are formed by appending terms that follow all existing terms in a given ordering. The formed tree is searched in a breadth-first manner. Given the nodes at each level, the dataset is scanned and the support is counted for each group. Each node is then examined to determine if it is significant and large, if it should be pruned, and if new children should be generated. After all significant contrast sets are located, a post-processor selects a subset to show to the user - the low order, simpler results are shown first, followed by the higher order results which are "surprising and significantly different." The support calculation comes from testing a null hypothesis that the contrast set support is equal across all groups (i.e., that contrast set support is independent of group membership). The support count for each group is a frequency value that can be analyzed in a contingency table where each row represents the truth value of the contrast set and each column variable indicates the group membership frequency. If there is a difference in proportions between the contrast set frequencies and those of the null hypothesis, the algorithm must then determine if the differences in proportions represent a relation between variables or if it can be attributed to random causes. This can be determined through a chi-square test comparing the observed frequency count to the expected count. Nodes are pruned from the tree when all specializations of the node can never lead to a significant and large contrast set. The decision to prune is based on: The minimum deviation size: The maximum difference between the support