Materials Property Prediction
materials science
Materials Property Prediction is the supervised machine learning task of mapping a material's fundamental crystal structure and chemical composition directly to its macroscopic…
27,049 technical terms and definitions
materials science
Materials Property Prediction is the supervised machine learning task of mapping a material's fundamental crystal structure and chemical composition directly to its macroscopic…
Materials Science: Structure, Defects, Phase Diagrams, and Mechanical Properties Materials science explains why a bridge alloy survives cyclic load, why a turbine blade retains…
materials science
Materials Science NLP is the application of natural language processing to extract structured knowledge from materials science literature — identifying material compositions…
reasoning, LLM, theorem, proving, symbolic, computation, verification
Math Reasoning LLM Theorem Proving is language models trained to perform mathematical reasoning, solve complex problems, and generate formal proofs, combining neural and symbolic…
math, evaluation
MATH is the competition-level mathematics benchmark of 12,500 problems drawn from AMC, AIME, and similar olympiad contests — designed to probe whether language models can perform…
math, evaluation
MATH Dataset is a challenging competition-level math benchmark covering advanced high-school problem solving - It is a core method in modern AI evaluation and safety execution…
architecture
Math Model is model specialization focused on formal reasoning, symbolic manipulation, and quantitative problem solving - It is a core method in modern semiconductor AI serving…
Mathematical Finance: Stochastic Calculus, Black–Scholes, and Risk-Neutral Pricing Mathematical finance studies uncertain cash flows using probability, optimization…
reasoning
Mathematical reasoning in AI involves solving mathematical problems through multi-step logical inference — including arithmetic, algebra, geometry, calculus, combinatorics…
mathematics science, mathematical modeling, semiconductor math, crystal growth math, czochralski equations, dopant segregation, heat transfer equations, lithography math
Mathematics Modeling 1. Crystal Growth (Czochralski Process) Growing single-crystal silicon ingots requires coupled models for heat transfer, fluid flow, and mass transport. 1.1…
evaluation
MathQA is the large-scale math word problem dataset annotated with executable operation programs — bridging the gap between end-to-end answer prediction and interpretable program…
pyplot, matplotlib figure, matplotlib axes, matplotlib animation, matplotlib scatter, matplotlib backends, matplotlib rcparams, matplotlib style, python visualization, matplotlib savefig, matplotlib subplots
Matplotlib is a Python plotting library that organizes every visualization as a hierarchy of Artist objects—Figure containing one or more Axes, each Axes containing Line2D, Patch…
matrix chain multiplication, matrix chain, chain multiplication
Matrix Chain Multiplication Introduction & Motivation Matrix Chain Multiplication addresses a central problem in algorithms process control and multiplication-order-optimization…
SoftImpute Nuclear-Norm Matrix Completion Introduction & Motivation SoftImpute Nuclear-Norm Matrix Completion addresses a central problem in {subject} assurance and {target}…
quality & reliability
Matrix Diagram is a cross-relationship chart that evaluates strength of linkage between two or more variable sets - It is a core method in modern semiconductor quality governance…
metrology
Matrix Effect in metrology is the influence of the sample composition (matrix) on the analytical signal of the target analyte — the same concentration of analyte can produce…
doe
Matrix experiments is the design-of-experiments method that varies multiple process factors simultaneously using structured test matrices - it reveals both main effects…
recommendation systems
Matrix factorization is a recommendation approach that decomposes user-item interaction matrices into latent user and item factors - Low-rank embeddings capture preference…
matrix factorization latent factor prediction, matrix factorization latent, factorization latent factor, latent factor prediction, matrix factorization, factorization latent
Matrix Factorization Latent Factor Prediction with Machine Learning Introduction & Motivation Matrix Factorization Latent Factor Prediction with Machine Learning addresses…
Matrix Factorization for Recommendations Introduction & Motivation Matrix Factorization for Recommendations addresses a central problem in recsys systems and {target}…
Singular Value Decomposition for Recommender Systems Introduction & Motivation Singular Value Decomposition for Recommender Systems addresses a central problem in recsys systems…
quality
Matrix-Matched Standard is a calibration standard prepared in the same matrix (background composition) as the sample being measured — ensuring the calibration standard experiences…
matmul, gemm, matrix multiply, tiled matrix multiplication, tensor core, systolic array
Matrix multiplication contracts the inner dimension of two matrices so C = A times B, with general work proportional to M times N times K. Dense matrix multiplication is…
matrix multiplication optimization, matrix multiplication, multiplication optimization
Matrix Multiplication Optimization Introduction & Motivation Matrix Multiplication Optimization addresses a central problem in ml-hardware process control and compute-utilization…
time series models
Matrix profile is a time-series primitive that stores nearest-neighbor distance for each subsequence in a series - Sliding-window similarity search identifies motifs discords…
matrix profile motif, matrix profile, profile motif
Matrix Profile Motif Introduction & Motivation Matrix Profile Motif addresses a central problem in timeseries process control and matrix-profile-subsequence-distance prediction…
Matrix Profile Discord Discovery for Time Series Introduction & Motivation Matrix Profile Discord Discovery for Time Series addresses a central problem in anomaly detection…
Matrix Profile for Exact Motif and Discord Mining Introduction & Motivation Matrix Profile for Exact Motif and Discord Mining addresses a central problem in timeseries forecasting…
Matrix-Variate Normal Distributions and Covariance Modeling Introduction & Motivation Matrix-Variate Normal Distributions and Covariance Modeling addresses a central problem…
rag
Matryoshka Embeddings is embeddings trained so truncated prefixes retain meaningful performance at multiple dimensional budgets - It is a core method in modern engineering…
production
Mature Yield is the steady-state yield achieved after the learning phase is complete — the maximum achievable yield for a given technology and product, limited by fundamental…
mawps, evaluation
MAWPS (Math Word Problem Repository) is the unified testbed for evaluating arithmetic word problem solvers — aggregating multiple elementary math datasets (AddSub, MultiArith…
ai agents
Max Iterations is a hard loop-count limit that prevents runaway reasoning and repetitive action cycles - It is a core method in modern semiconductor AI-agent engineering…
text generation
Max length is the hard upper bound on generated token count for a response or completion request - it is the primary guardrail against unbounded decoding. What Is Max length?…
structured prediction
Max-margin parsing is parsing methods that optimize structured margin objectives to separate correct and incorrect parses - Training emphasizes high-score separation for gold…
optimization
Max Tokens is an upper bound on generated token count to control latency, cost, and output size - It is a core method in modern semiconductor AI serving and inference-optimization…
maxcut optimization quantum, maxcut optimization, optimization quantum
MaxCut Optimization Quantum Introduction & Motivation MaxCut Optimization Quantum addresses a central problem in quantum process control and maxcut-graph-partition-problem…
Maximal Marginal Relevance MMR Diversity Re-Ranking Introduction & Motivation Maximal Marginal Relevance MMR Diversity Re-Ranking addresses a central problem in recsys systems…
graph algorithms
Maximum Common Subgraph (MCS) is the graph-theoretic problem of finding the largest subgraph that appears (up to isomorphism) as a subgraph of both input graphs simultaneously…
reinforcement learning
Maximum Entropy RL is a reinforcement learning framework that augments the standard reward with an entropy bonus — the agent maximizes the expected reward PLUS the entropy of its…
Soft Actor-Critic and Maximum Entropy RL Introduction & Motivation Soft Actor-Critic and Maximum Entropy RL addresses a central problem in reinforcement policy learning…
maximum flow min-cut prediction, maximum flow min-cut, flow min-cut prediction, maximum flow, flow min-cut, min-cut prediction
Maximum Flow Min-Cut Prediction with Machine Learning Introduction & Motivation Maximum Flow Min-Cut Prediction with Machine Learning addresses a central problem in maximum flow…
maximum flow minimum cut, maximum flow minimum, flow minimum cut, maximum flow, flow minimum, minimum cut
Maximum Flow Minimum Cut Introduction & Motivation Maximum Flow Minimum Cut addresses a central problem in algorithms process control and flow-value-optimization prediction: how…
mmd, domain adaptation
Maximum Mean Discrepancy (MMD) is a non-parametric statistical test and distance metric that measures the difference between two probability distributions by comparing their mean…
Maximum Mean Discrepancy MMD for Domain Alignment Introduction & Motivation Maximum Mean Discrepancy MMD for Domain Alignment addresses a central problem in transfer learning…
process
Maximum queue time is the hard upper limit on allowable waiting time between specified process steps before product quality risk becomes unacceptable - it enforces chemistry…
neural architecture
Maxout is a learnable activation function that takes the elementwise maximum of several linear transformations. Instead of using a fixed shape such as ReLU or sigmoid, Maxout lets…
maxq, reinforcement learning
MAXQ is a hierarchical RL value decomposition method that breaks down the overall value function into a sum of subtask completion rewards and subtask values — each node in…
reinforcement learning advanced
MAXQ Decomposition is value-function decomposition framework that breaks tasks into recursively defined subtasks. - It separates completion value and subtask value to support…
non-degenerate semiconductor approximation, thermal velocity kinetic theory, law of mass action np, einstein relation diffusion mobility, thermionic emission schottky barrier
Maxwell–Boltzmann Statistics: Non-Degenerate Carrier Transport, Thermal Velocity Distributions, and Semiconductor Kinetic Theory Executive Overview Maxwell–Boltzmann (MB)…