SARSA (State-Action-Reward-State-Action)
SARSA (State-Action-Reward-State-Action) is an on-policy temporal-difference (TD) control algorithm that learns the action-value function Q^pi(s, a) of the policy an agent is actually following
Machine Learning
Explore learning methods, model architectures, datasets, and practical applications.
Explore articles
Browse subtopics (63)
Articles that also belong to these categories. Counts cover all of Machine Learning.
All articles (858)Deep Learning (187)Natural Language Processing (92)Statistics (90)Training & Optimization (70)Data & Datasets (65)Neural Networks (58)Reinforcement Learning (58)Model Evaluation (55)Artificial Intelligence (54)Large Language Models (49)Mathematics (48)Computer Vision (41)AI Ethics (35)AI Infrastructure (29)AI Safety (26)Open Source AI (25)Data Science (24)AI Benchmarks (22)Developer Tools (20)Algorithms (18)AI Hardware (17)AI Tools & Products (17)Google (16)Generative AI (14)Information Retrieval (14)Model Architecture (13)People (13)AI Agents (12)Interpretability (12)Robotics (12)AI Companies (11)Transformer Models (11)AI Inference (10)MLOps (10)AI Models (8)Computer Science (8)AI Code Generation (7)AI Research (7)Multimodal AI (7)Speech & Audio AI (7)Chinese AI (6)AI Policy & Regulation (5)Meta AI (5)Programming Languages (5)AI History (4)AI for Science (4)Prompt Engineering (4)AI Alignment (3)AI Events (3)Reasoning Models (3)Embodied AI (2)Software Development (2)2025 in artificial intelligence (1)Anthropic (1)Diffusion Models (1)Drug Discovery (1)Education AI (1)Enterprise AI (1)Healthcare AI (1)Image Generation (1)NVIDIA (1)Small Language Models (1)World Models (1)
Showing 661-720 of 858 articles
SGLang
SGLang (short for Structured Generation Language) is a high-performance, open-source serving framework for large language models and multimodal models, originally developed at UC Berkeley's Sky Computing Lab…
SMOTE (Synthetic Minority Over-sampling Technique)
SMOTE (Synthetic Minority Over-sampling Technique) is a data preprocessing algorithm that fixes class imbalance by creating new synthetic minority-class examples through interpolation between existing minority…
STaR (Self-Taught Reasoner)
STaR (Self-Taught Reasoner) is a self-training method that teaches a large language model to reason by having it generate its own chain-of-thought rationales, keeping only the rationales that lead to correct…
SUPERB
SUPERB, which stands for Speech processing Universal PERformance Benchmark, is a comprehensive evaluation framework designed to measure how well self-supervised learning (SSL) models generalize across a…
Safetensors
Safetensors is an open-source tensor serialization format developed by Hugging Face that stores machine learning model weights as raw tensor data plus a small JSON header
Sam McCandlish
Sam McCandlish is an American AI researcher and a co-founder of Anthropic, the artificial intelligence company behind the Claude family of language models.
Sampling Bias
Sampling bias is a systematic error in statistics and machine learning that occurs when a sample is collected so that some members of the intended population have a higher or lower probability of being…
Sanjay Ghemawat
Sanjay Ghemawat (born 1966) is an American computer scientist and software engineer, best known for co-creating the core distributed-systems infrastructure that powered Google's rise, including the Google File…
Saver
In machine learning, a Saver is a utility or class that persists and restores the state of a model, its variables, and its optimizer to disk, so training can be paused, resumed, evaluated, or deployed without…
Scalar
A scalar is a single numerical value, a quantity with magnitude but no direction, and the simplest object in linear algebra and machine learning.
Scaling
In machine learning and artificial intelligence, scaling is an umbrella term for several distinct practices: (1) feature scaling, the data-preprocessing step that rescales input features to a common range so…
Scaling Laws
Neural scaling laws are empirical relationships that describe how a measured outcome changes as a neural network, its training data, or the computation used to train it becomes larger.
Scaling Laws for Neural Language Models
Scaling Laws for Neural Language Models is a landmark research paper published by OpenAI on January 23, 2020 (arXiv:2001.08361) that established that the test loss of a neural language model falls as a smooth…
Scikit-learn
Scikit-learn is a free and open-source machine learning library for Python. It provides a common interface for fitting, transforming, predicting, evaluating, and selecting models, with implementations covering…
Score matching
Score matching is a method for fitting a probabilistic model by matching the gradient of its log-density, the so-called score function $$\nabla_x \log p(x)$$, to the same gradient of the data distribution
Scoring
Scoring is the process of applying a trained machine learning model to data so it produces an output number, either a prediction (probability, class score, or ranking value) emitted at inference time, or a…
Selection Bias
Selection bias is a systematic error that occurs when the data used for analysis, training, or evaluation does not accurately represent the population or domain it is intended to describe
Selective Language Modeling (Rho-1)
Selective Language Modeling (SLM) is a pretraining objective for language models that applies the training loss to only a chosen subset of tokens rather than to every token in the corpus.
Self-RAG
Self-RAG (Self-Reflective Retrieval-Augmented Generation) is a framework that trains a single large language model to adaptively decide when to retrieve external passages, to generate text grounded in those…
Self-Supervised Learning
Self-supervised learning (SSL) is a machine learning approach in which a model learns representations from unlabeled data by generating its own supervisory signal from the structure of the data itself
Self-attention
Self-attention is a neural network operation in which each position forms a data-dependent mixture of information from positions in the same input sequence or set.
Self-speculative decoding (LayerSkip)
Self-speculative decoding is a family of speculative decoding methods that accelerate large language model (LLM) inference by using the target model itself, run in a cheaper reduced-depth mode
Self-training
Self-training is a semi-supervised learning procedure in which a model trained on a small labeled set is used to generate predictions on unlabeled data
Semantic search
Semantic search is an information retrieval approach that finds results based on the meaning and intent behind a query rather than relying solely on exact keyword matches.
Semi-Supervised Learning
Semi-supervised learning is a machine learning approach that trains a model on a small amount of labeled data together with a large amount of unlabeled data
Sensitive Attribute
A sensitive attribute (also called a protected attribute or protected characteristic) is any feature in a dataset that corresponds to a legally or ethically protected personal trait, such as race, sex or…
Sensor fusion
Sensor fusion is the process of combining data from multiple sensors, often of different types, to produce information that is more accurate, complete, or reliable than any single sensor could provide on its…
SentencePiece
SentencePiece is an open-source, language-agnostic subword tokenization library and detokenizer developed at Google and introduced by Taku Kudo and John Richardson in their 2018 EMNLP system demonstration…
Sentiment Analysis
Sentiment analysis, also called opinion mining or emotion AI, is the use of natural language processing to automatically determine the sentiment, emotion, or opinion expressed in a piece of text, typically…
Sequence Model
A sequence model is a class of machine learning models that processes, generates, or predicts ordered data where the position and surrounding context of each element carry meaning.
Sequence-to-Sequence Task
A sequence-to-sequence (seq2seq) task is any machine learning problem in which a model receives a variable-length input sequence and produces a variable-length output sequence, where the input and output may…
Serving
In machine learning, serving (also called model serving or inference serving) is the process of deploying a trained model into a production environment so it can answer prediction requests, usually exposed as…
Shape (Tensor)
In machine learning, the shape of a tensor is the tuple of integers giving the size of the tensor along each of its axes.
Shapley value
The Shapley value is a rule from cooperative game theory for dividing the payoff of a group among its members.
Shrinkage
Shrinkage in machine learning and statistics is a regularization technique that deliberately pulls model coefficient estimates toward zero, or toward some other fixed shrinkage target
Sigmoid Function
The sigmoid function is a mathematical function that maps any real number to a value between 0 and 1 using the formula $$\sigma(x) = \frac{1}{1 + e^{-x}}$$, producing a smooth S-shaped curve.
SimCLR
SimCLR (Simple Framework for Contrastive Learning of Visual Representations) is a self-supervised learning method for computer vision in which a network is trained to recognise that two differently augmented…
Similarity Measure
A similarity measure (also called a similarity function or similarity metric) is a real-valued function that quantifies the degree of resemblance between two objects, returning a large value when the objects…
Simulated annealing
Simulated annealing is a probabilistic method for finding a good approximate solution to a global optimization problem, especially one with many local optima that trap ordinary hill-climbing methods.
Singular value decomposition
Singular value decomposition (SVD) is a matrix factorization that writes any real or complex m x n matrix $$A$$ as the product $$A = U \Sigma V^\top$$, where U and V are orthogonal matrices (the left and right…
Sleep-time compute
Sleep-time compute is a technique for large language model inference in which a model uses idle periods, before any user query has arrived, to "think" about a known context offline and pre-compute a richer…
Small language model
A small language model (SLM) is a compact large language model, typically under about 10 billion parameters, built to run efficiently on a single consumer GPU, a laptop, or a smartphone rather than in a data…
Snorkel
Snorkel is an open-source software framework and methodology for programmatic data labeling that started at Stanford University in 2015 and is licensed under Apache 2.0 .
Social Media
Social media comprises internet-based services through which people and organizations create, distribute, discover, and respond to content in networked publics.
Softmax
Softmax is a function that converts a finite vector of real-valued scores into a vector of positive numbers that sum to one.
Sparse Coding
Sparse coding is a representation learning principle in which a signal is encoded as a linear combination of a small number of elements drawn from a (usually overcomplete) dictionary of atoms.
Sparse Feature
A sparse feature is a feature in machine learning whose values are predominantly zero or empty across a dataset.
Sparse Representation
Sparse representation is a data encoding strategy in which most values in a vector, matrix, or tensor are zero (or near zero), with only a small fraction of elements carrying non-zero values.
Sparse Vector
A sparse vector is a vector in which most of the elements are zero, in contrast to a dense vector, in which most elements are non-zero.
Sparse attention
Sparse attention is a family of techniques that cut the computational and memory cost of the attention mechanism in transformer models by letting each token attend to only a subset of other tokens in a sequence
Sparse autoencoder
A sparse autoencoder (SAE) is a neural network that adds a sparsity penalty to an autoencoder's training loss so that only a small number of hidden units activate for any given input, producing a wide
Sparse upcycling
Sparse upcycling is a technique for building a sparsely activated mixture of experts (MoE) model by initializing it from an already trained dense Transformer checkpoint
Sparsity
Sparsity is the property of a data structure, model, or representation in which most values are zero or near-zero
Spatial Pooling
Spatial pooling is a downsampling operation in convolutional neural networks (CNNs) that replaces a local region of a feature map with a single summary statistic, such as the maximum or the average of the…
Speculative Decoding
Speculative decoding is a lossless inference acceleration technique for autoregressive transformer models in which a small, fast draft model proposes several future tokens at once and the larger target model…
Speech Recognition
Speech recognition, usually called automatic speech recognition (ASR), is the computational task of converting a spoken-language signal into a sequence of written symbols.
SpinQuant
SpinQuant is a post-training quantization method for large language models that inserts learned rotation matrices into a transformer network to make its weights, activations, and KV cache easier to represent…
Split
A split in machine learning is a partitioning operation, and the word names two distinct things.
Splitter
A splitter is a term used in two distinct senses in machine learning. The first and most common sense is a utility that partitions a dataset into subsets such as training, validation, and test sets, or into…