Biomedical data science is a multidisciplinary field which leverages large volumes of data to promote biomedical innovation and discovery. Biomedical data science draws from various fields including Biostatistics, Biomedical informatics, and machine learning, with the goal of understanding biological and medical data. It can be viewed as the study and application of data science to solve biomedical problems. Modern biomedical datasets often have specific features which make their analyses difficult, including: Large numbers of feature (sometimes billions), typically far larger than the number of samples (typically tens or hundreds) Noisy and missing data Privacy concerns (e.g., electronic health record confidentiality) Requirement of interpretability from decision makers and regulatory bodies Many biomedical data science projects apply machine learning to such datasets. These characteristics, while also present in many data science applications more generally, make biomedical data science a specific field. Examples of biomedical data science research include: Computational genomics Computational imaging Electronic health records data mining Biomedical network science Clinical Natural Language Processing (NLP) == Computational Imaging and Deep Learning == Computational imaging is a cornerstone of biomedical data science, focusing on the development of algorithms to enhance, analyze, and interpret medical imagery. In recent years, the field has been transformed by the integration of deep learning, particularly through the use of Convolutional Neural Networks. Deep learning started from researchers manually defining characteristics like edge detection or texture representation learning. In a more modern approach of computational imaging, models automatically learn a hierarchy of features directly from raw pixel data. This overlap between data science and deep learning is applied across several key tasks: Classification: Identifying the presence of specific diseases, such as distinguishing between benign and malignant tumors in histopathology slides or detecting pneumonia in chest X-rays. Segmentation: The precise delineation of anatomical structures or lesions. A notable example is the U-Net architecture, which is widely used for biomedical image segmentation to help clinicians quantify organ volume or track tumor growth. Detection: Automating the localization of small objects, such as identifying microcalcifications in mammograms or polyps during colonoscopies. Registration: The process of aligning multiple images to provide a comprehensive view of the patient's anatomy. Even with all of these enhancements, the application of deep learning in medical imaging requires accomplishing vigorous challenges. An example of these changes is building large, annotated datasets and creating the imperative for model interpretability in clinical decision-making. == Electronic Health Records == Electronic Health Records (EHRs) are a digital alternative to patient paper charts, usually including individual records or population health information. EHRs can be used in a wide variety of applications, including research and analysation as they often include demographics, diagnoses, medications, test results, and personal statistics. === History === ==== 1960s ==== The earliest precursor is considered Dr. Lawrence Weed's problem-oriented medical record (POMR) published in the 1968 which sorts and groups medical records by medical diagnoses and symptoms. The POMR was the first system to organize based off of patient information rather than the source (doctors, nurses, attendings, etc.). In 1969, the Regenstrief Institute developed and published the Regenstrief Medical Record System which established electronic writing, storage, and retrieval of records which served as the basis for modern EHR systems. ==== 2000s ==== In 2009, the Health Information Technology for Economic and Clinical Health Act (HITECH Act) was passed in the United States. This act standardized privacy and distribution of EHRs and increased the acceptance and utilization of EHRs within medical and academic settings. == Artificial Intelligence and Machine Learning Applications == Machine Learning and Artificial Intelligence have become central tools in biomedical data science. Recent advances in large language models (LLMs) have expanded their role beyond text, with models trained directly on genomic sequences enabling tasks such as gene function prediction, variant effect analysis, and drug discovery. In clinical settings, Natural Language Processing (NLP) models are applied to electronic health records to extract structured insights from unstructured clinical notes and data, supporting diagnosis and treatment planning. Beyond genomics, AI models have been applied to protein structure prediction. AlphaFold, developed by Google DeepMind, uses deep learning to predict three-dimensional protein structures from amino acid sequences with high accuracy. These predictions have been used to support drug target identification and the study of disease mechanisms. == Knowledge Graphs == Knowledge graphs (KGs) are widely used in biomedical data science to represent and analyze complex relationships among biological and medical entities. By structuring data as nodes (e.g., genes, diseases, drugs) and edges (relationships), KGs enable computational methods to extract insights and support decision-making. These biomedical relationships can be efficiently modeled and queried using technologies such as Neo4j. === Biomedical Research Applications === KGs provide biomedical researchers with a way to model complex biological systems. They have been used to identify the relationships between diseases and biomolecules, support drug repurposing, and to uncover new biological insights. Additional applications include: Identification of novel antibiotic resistance genes through graph-based link prediction. Finding associations between miRNA and diseases. Prediction of protein-protein interactions. === Clinical Applications === In clinical settings, KGs can be used to make visual representations of a patient's electronic health records. The data obtained from these graphs can assist healthcare providers in improving patient diagnoses and prescribing more effective drugs. Additionally, embeddings derived from resources like the Unified Medical Language System (UMLS) enable natural language processing of clinical text and similarity analysis between medical concepts. === Limitations === Despite their advantages, knowledge graphs face several challenges. Some of these include: High algorithmic complexity and large biological datasets make the process computationally expensive. KG construction can be a time-consuming process that requires careful attention to assign appropriate node types and vocabularies. Using data from a wide range of datasets in one KG requires them to be effectively integrated. == Privacy == A primary challenge in biomedical data science is maintaining medical privacy. Conducting research requires that data be collected on a number of people for training and testing purposes and is stored within biomedical datasets. This poses a risk for violating patient confidentiality and may dissuade people from participating in studies. The main sources of health statistics are surveys administrative and medical records health care claims data, vital records surveillance disease registries grey literature and peer-reviewed literature. Large data collection is a useful tool for researching various medical conditions. Researchers use these large datasets of information to identify factors that may make people more susceptible to certain diseases. Large amounts of collected data can help researchers identify patterns for disease probabilities. The findings can show a person is more likely for a condition, or identify environmental, social, and personal habits that may lead to adverse health issues. Institutions researching using personal medical information come with a moral and legal responsibility to protect the use of that information. Protection of the collected information has become a big concern. Sophisticated and coordinated attacks on certain medical systems happen more frequently. Medical companies, medical insurance and private businesses have invested a great deal into the protection of personal data. Despite this, data breaches continue to be documented. The chart below shows the top healthcare breaches in 2025. For these reasons, many people have reservations about giving up their personal data. Aside from the legitimate use of personal data there have been instances where companies have found methods to profit from brokering medical information. Concerns exist regarding unauthorized use of sensitive information within these data companies. If a person is identified within a dataset, then sensitive data can be used to discriminate against them. For example, insurance companies may charge a hi
Ed (chatbot)
Ed was a chatbot co-developed by the Los Angeles Unified School District and AllHere Education. Described as a learning acceleration platform, it was the first personal assistant for students in the United States. Part of the district's Individual Acceleration Plan, it was able to interact with students both verbally and visually, offering support in 100 languages. The chatbot was launched on March 20, 2024, as part of the district's plan for academic recovery from the COVID-19 pandemic and to improve overall academic performance. Utilizing artificial intelligence, Ed organizes data and reports on grades, test scores, and attendance, creating individualized plans for each student. After the company behind it, AllHere, collapsed, the district shuttered operations of the chatbot on June 14, 2024. The firm is under investigation by the US Federal Bureau of Investigation. == History == On February 14, 2022, Alberto M. Carvalho became the Superintendent of the Los Angeles Unified School District, pledging to give the district a full academic recovery from the COVID-19 pandemic. In December 2022, he announced the Individual Acceleration Plan for the district, which aimed to provide each student with a unique progress report and help them determine if they were on track to graduate. The district faced criticism from disability advocates for its management of Individualized Education Programs, and in April 2022, the United States Department of Education announced that the district had failed to provide appropriate educational services to students with disabilities during the pandemic. The district had been grappling with significant absenteeism issues since the pandemic, which led to declining academic performance and disengagement among students. On February 17, 2023, the district issued a request for proposals to develop a fully integrated portal system. Later that year, they signed a $6 million, five-year contract with AllHere Education, a Boston-based company founded in 2016. The introduction of Ed follows the public launch of ChatGPT, which has been utilized by both teachers and students in educational settings. On August 4, 2023, during an annual address at the Walt Disney Concert Hall, Carvalho and the Los Angeles Unified School District announced the launch of Ed. The district invested $4 million into the chatbot, with Carvalho noting that this cost would be halved thanks to donor and grant funding. The chatbot was launched on March 20, 2024. Following its launch, a press conference was held to address security and technology concerns. Carvalho stated that the district had collaborated with security companies and incorporated filters to screen for threatening language. Months after its launch, AllHere Education furloughed most of its staff on June 14, citing their “current financial position” on its website as the reason. After learning about the furlough, the district terminated its dealings with AllHere Education. However, it stated its intention to bring the chatbot back in the future once officials determine the best course of action. Carvalho announced that he would appoint an independent task force to review what went wrong with AllHere Education and the chatbot. On February 25, 2026, the FBI served a search warrant on Carvalho’s home and office in connection with AllHere. The FBI also raided the LAUSD's headquarters. == Service == The chatbot was described as a personal assistant and a "one-stop shop for parents and students" who want to see information about a student's attendance and grades, as well as other resources from the district. Additionally, the application can function as an alarm clock, provide daily lunch menus from the school cafeteria, and offer updates on the location of school buses. The chatbot also helps students and parents who do not speak English as their first language by translating displayed information into approximately 100 different languages. The application can also help with submitting applications and give updates on progress and upcoming assignments. The district stated that the primary goal of Ed was to actively motivate students to complete homework and other tasks. == Reception == The chatbot received a mostly positive reception among parents and observers upon its launch. Some parents and teachers expressed caution about the technology, voicing concerns that the district's push for its implementation lacked public accountability. Rob Nelson from the University of Pennsylvania described the district's strategy as risky, saying that the release felt "like the beginning of a Clippy-level disaster". After the chatbot's shutdown, The 74 criticized it for misusing student data. Chris Whiteley, a former software engineer at AllHere Education, alleged that the data collected by the chatbot likely violated the district's data privacy rules.
Weibo (Chinese: 微博; pinyin: Wēibó), or Sina Weibo (Chinese: 新浪微博; pinyin: Xīnlàng Wēibó), is a Chinese microblogging (weibo) website. Launched by Sina Corporation on 14 August 2009, it is one of the biggest social media platforms in China, with over 582 million monthly active users (252 million daily active users) as of Q1 2022. The platform has been highly successful but has faced criticism for heavy censorship. Sina had gone public on the Nasdaq in 2000. In March 2014, Sina announced a spinoff of Weibo and filed an IPO under the symbol WB. Sina carved out 11% of Weibo in the IPO, with Alibaba owning 32% post-IPO. The company began trading publicly on 17 April 2014. In March 2017, Sina launched Sina Weibo International Version. In November 2018, Sina Weibo suspended its registration function for minors under the age of 14. In July 2019, Sina Weibo announced that it would launch a two-month campaign to clean up pornographic and vulgar information, named "Project Deep Blue" (蔚蓝计划). On 29 September 2020, the company announced it would go private again due to rising tensions between the US and China. == Name == "Weibo" (微博) is the Chinese word for "microblog". Sina Weibo launched its new domain name weibo.com on 7 April 2011, deactivating and redirecting from the old domain, t.sina.com.cn, to the new one. Due to its popularity, the media sometimes refers to the platform simply as "Weibo", despite the numerous other Chinese microblogging services including Tencent Weibo, Sohu Weibo, and NetEase Weibo. However, the latter three have stopped providing services. == Background == Sina Weibo is a platform based on fostering user relationships to share, disseminate, and receive information. Through the website or the mobile app, users can upload pictures and videos publicly for instant sharing, with other users being able to comment with text, pictures and videos, or use a multimedia instant messaging service. The company initially invited a large number of celebrities to join the platform at the beginning and has since invited many media personalities, government departments, businesses and non-governmental organizations to open accounts for the purpose of publishing and communicating information. To avoid the impersonation of celebrities, Sina Weibo uses verification symbols; celebrity accounts have an orange letter "V" and organizations' accounts have a blue letter "V". Sina Weibo has more than 500 million registered users; out of these, 313 million are monthly active users, 85% use the Weibo mobile app, 70% are college-aged, 50.10% are male and 49.90% are female. There are over 100 million messages posted by users each day. With more than 100 million followers, actress Xie Na holds the record for the most followers on the platform. Despite fierce competition among Chinese social media platforms, Sina Weibo remains the most popular. == History == After the July 2009 Ürümqi riots, China shut down most domestic microblogging services, including Fanfou, the very first weibo service. Many popular non-China-based microblogging services like Twitter, Facebook, and Plurk have since been blocked. Sina Corporation CEO Charles Chao considered this to be an opportunity, and on 14 August 2009, Sina launched the tested version of Sina Weibo. Basic functions including message, private message, comment and reposting were made available that September. A Sina Weibo–compatible API platform for developing third-party applications was launched on 28 July 2010. On 1 December 2010, the website experienced an outage, which administrators later said was due to the ever-increasing numbers of users and posts. Registered users surpassed 100 million in February 2011. Since 23 March 2011, t.cn has been used as Sina Weibo's official shortened URL in lieu of sinaurl.cn. On 7 April 2011, weibo.com replaced t.sina.com.cn as the new main domain name used by the website. The official logo was also updated. In June 2011, Sina announced an English-language version of Sina Weibo would be developed and launched, though content would still be governed by Chinese law. On 11 January 2013, Sina Weibo and Alibaba China (a subsidiary of Alibaba Group) signed a strategic cooperation agreement. With more and more foreign celebrities using Sina Weibo, language translation has become an urgent need for Chinese users who wish to communicate with their idols online, especially Korean. In January 2013, Sina Weibo and NetEase.com announced that they had reached a strategic cooperation agreement. When users browse foreign language content, they can now directly obtain translation results through the YouDao Dictionary. The Sina Weibo financial report in February 2013 showed that its total revenue was approximately US$66 million and that the number of registered users had exceeded the 500 million mark. In April 2013, Sina officially announced that Sina Weibo had signed a strategic cooperation agreement with Alibaba. The two sides conducted in-depth cooperation in areas such as user account interoperability, data exchange, online payment, and internet marketing. At the same time, Sina announced that Alibaba, through its wholly owned subsidiary, had purchased the preferred shares and common shares issued by Sina Weibo Company for US$586 million, which accounted for approximately 18% of Weibo's fully diluted and diluted total shares. === Ownership === On 9 April 2013, Alibaba Group announced that it would acquire 18% of Sina Weibo for US$586 million, with the option to buy up to 30% in the future. Alibaba exercised this option when Weibo was listed on the NASDAQ in April 2014. == Users == According to iResearch's report on 30 March 2011, Sina Weibo had 56.5% of China's microblogging market based on active users and 86.6% based on browsing time over competitors such as Tencent Weibo and Baidu. According to research by Sina Corporation, the number of active users reached over 400 million by Q1 2018, making Sina Weibo the 7th platform with at least 400 million active users, and daily usage increased by 21%. As of 2017, approximately 80% of its users were in their 20s and 30s. The top 100 users had over 485 million followers combined. More than 5,000 companies and 2,700 media organizations in China use Sina Weibo. The site is maintained by a growing microblogging department of 200 employees responsible for technology, design, operations, and marketing. Sina executives invited and persuaded many Chinese celebrities to join the platform. Users now include Asian celebrities, movie stars, singers, famous business and media figures, athletes, scholars, artists, organizations, religious figures, government departments, and officials from Hong Kong, Mainland China, Malaysia, Singapore, Taiwan, and Macau, as well as some famous foreign individuals and organizations, including Kevin Rudd, Boris Johnson, David Cameron, Narendra Modi, Toshiba, and the Germany national football team. Sina Weibo has a verification program for known people and organizations. Once an account is verified, a verification badge is added beside the account name. == Features == Many of Sina Weibo's features resemble those of Twitter. A user may post with a 140-character limit (increased to 2,000 as of January 2016 with the exception of reposts and comments). An analysis of 29 million Weibo posts found the median length was 14 characters. Users may mention or talk to other people using "@UserName" formatting, add hashtags, follow other users to make their posts appear in one's own timeline, re-post with "//@UserName" similar to Twitter's retweet function "RT @UserName", select posts for one's favorites list, and verify the account if the user is a celebrity, brand, business or otherwise of public interest. URLs are automatically shortened using the domain name t.cn, akin to Twitter's t.co. Official and third-party applications can access Sina Weibo from other websites or platforms. Users may: Submit up to 18 images/video files in every post Send personal messages to followers Follow others and be followed Post "stories" like on Instagram React to posts using different emojis Receive monetary rewards that can be used in a digital store linked to Weibo View posts identified as "hot" or popular Display the location they post from Hashtags differ slightly between Sina Weibo and Twitter, using the double-hashtag "#HashName#" format (the lack of spacing between Chinese characters necessitates a closing tag). Users can own a hashtag by requesting hashtag monitoring; the company reviews these requests and responds within one to three days. Once a user owns a hashtag, they have access to a wide variety of functions available only to them on the condition that they remain active (less than 1 post per calendar week revokes these privileges). Additionally, comments appear as a list below each post. A commenter can also choose to re-post the comment, quoting the whole original post, to their own page. Unregistered users can only browse a few post
Information schema
In relational databases, the information schema (information_schema) is an ANSI-standard set of read-only views that provide information about all of the tables, views, columns, and procedures in a database. It can be used as a source of the information that some databases make available through non-standard commands, such as: the SHOW command of MySQL the DESCRIBE command of Oracle's SQLPlus the \d command in psql (PostgreSQL's default command-line program). => SELECT count(table_name) FROM information_schema.tables; count ------- 99 (1 row) => SELECT column_name, data_type, column_default, is_nullable FROM information_schema.columns WHERE table_name='alpha'; column_name | data_type | column_default | is_nullable -------------+-----------+----------------+------------- foo | integer | | YES bar | character | | YES (2 rows) => SELECT FROM information_schema.information_schema_catalog_name; catalog_name -------------- johnd (1 row) == Implementation == As a notable exception among major database systems, Oracle does not as of 2015 implement the information schema. An open-source project exists to address this. RDBMSs that support information_schema include: Amazon Redshift Apache Hive Microsoft SQL Server MonetDB Snowflake MySQL PostgreSQL H2 Database HSQLDB InterSystems Caché MariaDB SingleStore (formerly MemSQL) Mimer SQL Snowflake Trino Presto CrateDB ClickHouse CockroachDB Kinetica DB TiDB RDBMSs that do not support information_schema include: Apache Derby Apache Ignite Firebird Microsoft Access IBM Informix Ingres IBM Db2 Oracle Database SAP HANA SQLite Sybase ASE Sybase SQL Anywhere Teradata Vertica
Information schema
In relational databases, the information schema (information_schema) is an ANSI-standard set of read-only views that provide information about all of the tables, views, columns, and procedures in a database. It can be used as a source of the information that some databases make available through non-standard commands, such as: the SHOW command of MySQL the DESCRIBE command of Oracle's SQLPlus the \d command in psql (PostgreSQL's default command-line program). => SELECT count(table_name) FROM information_schema.tables; count ------- 99 (1 row) => SELECT column_name, data_type, column_default, is_nullable FROM information_schema.columns WHERE table_name='alpha'; column_name | data_type | column_default | is_nullable -------------+-----------+----------------+------------- foo | integer | | YES bar | character | | YES (2 rows) => SELECT FROM information_schema.information_schema_catalog_name; catalog_name -------------- johnd (1 row) == Implementation == As a notable exception among major database systems, Oracle does not as of 2015 implement the information schema. An open-source project exists to address this. RDBMSs that support information_schema include: Amazon Redshift Apache Hive Microsoft SQL Server MonetDB Snowflake MySQL PostgreSQL H2 Database HSQLDB InterSystems Caché MariaDB SingleStore (formerly MemSQL) Mimer SQL Snowflake Trino Presto CrateDB ClickHouse CockroachDB Kinetica DB TiDB RDBMSs that do not support information_schema include: Apache Derby Apache Ignite Firebird Microsoft Access IBM Informix Ingres IBM Db2 Oracle Database SAP HANA SQLite Sybase ASE Sybase SQL Anywhere Teradata Vertica
Exploration–exploitation dilemma
The exploration–exploitation dilemma, also known as the explore–exploit tradeoff, is a fundamental concept in decision-making that arises in many domains. It is depicted as the balancing act between two opposing strategies. Exploitation involves choosing the best option based on current knowledge of the system (which may be incomplete or misleading), while exploration involves trying out new options that may lead to better outcomes in the future at the expense of an exploitation opportunity. Finding the optimal balance between these two strategies is a crucial challenge in many decision-making problems whose goal is to maximize long-term benefits. == Application in machine learning == In the context of machine learning, the exploration–exploitation tradeoff is fundamental in reinforcement learning (RL), a type of machine learning that involves training agents to make decisions based on feedback from the environment. Crucially, this feedback may be incomplete or delayed. The agent must decide whether to exploit the current best-known policy or explore new policies to improve its performance. === Multi-armed bandit methods === The multi-armed bandit (MAB) problem was a classic example of the tradeoff, and many methods were developed for it, such as epsilon-greedy, Thompson sampling, and the upper confidence bound (UCB). See the page on MAB for details. In more complex RL situations than the MAB problem, the agent can treat each choice as a MAB, where the payoff is the expected future reward. For example, if the agent performs an epsilon-greedy method, then the agent will often "pull the best lever" by picking the action that had the best predicted expected reward (exploit). However, it would pick a random action with probability epsilon (explore). Monte Carlo tree search, for example, uses a variant of the UCB method. === Exploration problems === There are some problems that make exploration difficult. Sparse reward. If rewards occur only once a long while, then the agent might not persist in exploring. Furthermore, if the space of actions is large, then the sparse reward would mean the agent would not be guided by the reward to find a good direction for deeper exploration. A standard example is Montezuma's Revenge. Deceptive reward. If some early actions give immediate small reward, but other actions give later large reward, then the agent might be lured away from exploring the other actions. Noisy TV problem. If certain observations are irreducibly noisy (such as a television showing random images), then the agent might be trapped exploring those observations (watching the television). === Exploration reward === This section based on. The exploration reward (also called exploration bonus) methods convert the exploration-exploitation dilemma into a balance of exploitations. That is, instead of trying to get the agent to balance exploration and exploitation, exploration is simply treated as another form of exploitation, and the agent simply attempts to maximize the sum of rewards from exploration and exploitation. The exploration reward can be treated as a form of intrinsic reward. We write these as r t i , r t e {\displaystyle r_{t}^{i},r_{t}^{e}} , meaning the intrinsic and extrinsic rewards at time step t {\displaystyle t} . However, exploration reward is different from exploitation in two regards: The reward of exploitation is not freely chosen, but given by the environment, but the reward of exploration may be picked freely. Indeed, there are many different ways to design r t i {\displaystyle r_{t}^{i}} described below. The reward of exploitation is usually stationary (i.e. the same action in the same state gives the same reward), but the reward of exploration is non-stationary (i.e. the same action in the same state should give less and less reward). Count-based exploration uses N n ( s ) {\displaystyle N_{n}(s)} , the number of visits to a state s {\displaystyle s} during the time-steps 1 : n {\displaystyle 1:n} , to calculate the exploration reward. This is only possible in small and discrete state space. Density-based exploration extends count-based exploration by using a density model ρ n ( s ) {\displaystyle \rho _{n}(s)} . The idea is that, if a state has been visited, then nearby states are also partly-visited. In maximum entropy exploration, the entropy of the agent's policy π {\displaystyle \pi } is included as a term in the intrinsic reward. That is, r t i = − ∑ a π ( a | s t ) ln π ( a | s t ) + ⋯ {\displaystyle r_{t}^{i}=-\sum _{a}\pi (a|s_{t})\ln \pi (a|s_{t})+\cdots } . === Prediction-based === This section based on. The forward dynamics model is a function for predicting the next state based on the current state and the current action: f : ( s t , a t ) ↦ s t + 1 {\displaystyle f:(s_{t},a_{t})\mapsto s_{t+1}} . The forward dynamics model is trained as the agent plays. The model becomes better at predicting state transition for state-action pairs that had been done many times. A forward dynamics model can define an exploration reward by r t i = ‖ f ( s t , a t ) − s t + 1 ‖ 2 2 {\displaystyle r_{t}^{i}=\|f(s_{t},a_{t})-s_{t+1}\|_{2}^{2}} . That is, the reward is the squared-error of the prediction compared to reality. This rewards the agent to perform state-action pairs that had not been done many times. This is however susceptible to the noisy TV problem. Dynamics model can be run in latent space. That is, r t i = ‖ f ( s t , a t ) − ϕ ( s t + 1 ) ‖ 2 2 {\displaystyle r_{t}^{i}=\|f(s_{t},a_{t})-\phi (s_{t+1})\|_{2}^{2}} for some featurizer ϕ {\displaystyle \phi } . The featurizer can be the identity function (i.e. ϕ ( x ) = x {\displaystyle \phi (x)=x} ), randomly generated, the encoder-half of a variational autoencoder, etc. A good featurizer improves forward dynamics exploration. The Intrinsic Curiosity Module (ICM) method trains simultaneously a forward dynamics model and a featurizer. The featurizer is trained by an inverse dynamics model, which is a function for predicting the current action based on the features of the current and the next state: g : ( ϕ ( s t ) , ϕ ( s t + 1 ) ) ↦ a t {\displaystyle g:(\phi (s_{t}),\phi (s_{t+1}))\mapsto a_{t}} . By optimizing the inverse dynamics, both the inverse dynamics model and the featurizer are improved. Then, the improved featurizer improves the forward dynamics model, which improves the exploration of the agent. Random Network Distillation (RND) method attempts to solve this problem by teacher–student distillation. Instead of a forward dynamics model, it has two models f , f ′ {\displaystyle f,f'} . The f ′ {\displaystyle f'} teacher model is fixed, and the f {\displaystyle f} student model is trained to minimize ‖ f ( s ) − f ′ ( s ) ‖ 2 2 {\displaystyle \|f(s)-f'(s)\|_{2}^{2}} on states s {\displaystyle s} . As a state is visited more and more, the student network becomes better at predicting the teacher. Meanwhile, the prediction error is also an exploration reward for the agent, and so the agent learns to perform actions that result in higher prediction error. Thus, we have a student network attempting to minimize the prediction error, while the agent attempting to maximize it, resulting in exploration. The states are normalized by subtracting a running average and dividing a running variance, which is necessary since the teacher model is frozen. The rewards are normalized by dividing with a running variance. Exploration by disagreement trains an ensemble of forward dynamics models, each on a random subset of all ( s t , a t , s t + 1 ) {\displaystyle (s_{t},a_{t},s_{t+1})} tuples. The exploration reward is the variance of the models' predictions. === Noise === For neural network–based agents, the NoisyNet method changes some of its neural network modules by noisy versions. That is, some network parameters are random variables from a probability distribution. The parameters of the distribution are themselves learnable. For example, in a linear layer y = W x + b {\displaystyle y=Wx+b} , both W , b {\displaystyle W,b} are sampled from Gaussian distributions N ( μ W , Σ W ) , N ( μ b , Σ b ) {\displaystyle {\mathcal {N}}(\mu _{W},\Sigma _{W}),{\mathcal {N}}(\mu _{b},\Sigma _{b})} at every step, and the parameters μ W , Σ W , μ b , Σ b {\displaystyle \mu _{W},\Sigma _{W},\mu _{b},\Sigma _{b}} are learned via the reparameterization trick.
Depth peeling
In computer graphics, depth peeling is an exact multipass method of order-independent transparency that extracts transparent fragments into depth layers and composites those layers in depth order. Depth peeling has the advantage of being able to generate correct results even for complex images containing intersecting transparent objects. == Method == Depth peeling works by rendering the image multiple times. Depth peeling uses two Z buffers, one that works conventionally, and one that is not modified, and sets the minimum distance at which a fragment can be drawn without being discarded. For each pass, the previous pass' conventional Z-buffer is used as the minimal Z-buffer, so each pass removes already-captured nearer fragments and draws the next depth layer behind them. The resulting images can then be composited in depth order to form a single image. A major drawback of classical depth peeling is performance: it requires one geometry pass per peeled layer, so scenes with high depth complexity require many passes that each re-rasterize the transparent geometry. Later variants reduce the number of passes by peeling multiple layers or both front and back layers in a pass. Dual depth peeling reduces the geometry-pass count from N to N/2+1 by peeling one layer from the front and one from the back in each pass, while multi-layer depth peeling peels several layers per pass and reported up to an 8x speed-up in RGBA8 settings.