collaborative-metric-learning-cml machine learning
Collaborative Metric Learning in Euclidean Space Introduction & Motivation Collaborative Metric Learning in Euclidean Space addresses a central problem in recsys systems…
2,848 technical terms and definitions
Collaborative Metric Learning in Euclidean Space Introduction & Motivation Collaborative Metric Learning in Euclidean Space addresses a central problem in recsys systems…
supply chain & logistics
Collaborative Planning is joint planning process across partners to align demand, supply, and execution assumptions - It reduces bullwhip effects and improves synchronized…
self-supervised learning
Collapse prevention in self-supervised learning is the set of architectural and loss-level constraints that stop encoders from mapping all inputs to identical or low-information…
allreduce, nccl
NCCL (NVIDIA Collective Communications Library) provides highly optimized GPU-to-GPU communication primitives for multi-GPU and multi-node deep learning training, with AllReduce…
Collective Sequence Anomaly Detection via HMMs Introduction & Motivation Collective Sequence Anomaly Detection via HMMs addresses a central problem in anomaly detection…
two level allreduce, node leader collectives, multi tier communication
Hierarchical Collective Communication is the multi-tier communication strategy that exploits the bandwidth and latency asymmetry of modern clusters by performing separate…
allreduce broadcast gather, nccl collective gpu, ring allreduce tree allreduce, communication primitive parallel
Collective Communication Primitives are the fundamental parallel communication operations that coordinate data exchange among all processes (or GPUs) in a group — where allreduce…
mpi collective operations, allreduce allgather broadcast, collective algorithm design, communication primitive optimization
Collective Communication Optimization is the algorithmic and systems-level techniques for efficiently implementing many-to-many communication patterns (all-reduce, all-gather…
allreduce broadcast gather, mpi collective operations, ring allreduce algorithm, communication topology patterns
Collective Communication Primitives — Fundamental building blocks for coordinating data exchange among multiple processes in parallel and distributed computing systems. Primary…
collective die placement accuracy prediction machine learning, collective die, placement accuracy prediction, rdl, collective die placement accuracy prediction
Collective Die Placement Accuracy Prediction with Machine Learning Introduction & Motivation Collective Die Placement Accuracy Prediction with Machine Learning addresses a central…
mpi io, parallel file read, hdf5 parallel write, aggregated io
Collective I/O (MPI-IO) is the coordinated parallel file access technique where multiple MPI processes cooperate to read or write a shared file in large, contiguous I/O operations…
transistor collector, collector current, collector-base junction, collector dissipation, heat sink tab, TO-220, BJT terminal
The collector is the terminal and region of a bipolar junction transistor that collects the carriers sent across the base, delivering the large output current of the device. It is…
Collider Stratification Bias and M-Bias Detection Introduction & Motivation Collider Stratification Bias and M-Bias Detection addresses a central problem in causal inference…
collimator, collimated pvd, collimated sputtering, long-throw pvd, long throw sputtering, directional pvd, pvd
A collimator is a filter that throws away most of what it is given, and the reason a fab would accept that is that a sputter target emits into the wrong shape. Atoms leaving…
manufacturing operations
Color Coding is using standardized colors to classify status, priority, or category in operations - It accelerates recognition and reduces interpretation errors. What Is Color…
self-supervised learning
Colorization as Pretext is a self-supervised learning task where the model is trained to predict the color channels (a, b in Lab color space) of an image given only the luminance…
Split-Brain Autoencoders and Image Colorization Pretext Introduction & Motivation Split-Brain Autoencoders and Image Colorization Pretext addresses a central problem…
self-supervised learning
Colorization pretext learning is the self-supervised task that predicts color channels from grayscale input so the model must infer semantic object identity and material cues…
Image Colorization Pretext Task Introduction & Motivation Image Colorization Pretext Task addresses a central problem in selfsupervised and color-space-selection architecture: how…
distributed training
Colossal-AI is the distributed training framework that unifies multiple parallelism strategies with automation for large-model optimization - it combines data, tensor…
bletchley park colossus, tommy flowers colossus, lorenz cipher codebreaking computer, first programmable electronic digital computer
Colossus was built in 1943 at Bletchley Park to solve a problem no previous machine in this history had faced: breaking a cipher fast enough to matter before the intelligence went…
coma, reinforcement learning advanced
COMA is counterfactual multi-agent policy gradients that compute agent-specific advantages with centralized critics - Counterfactual baselines estimate how each agent action…
yield enhancement
Comb-Serpentine is a paired monitor structure used to detect interconnect shorts and opens in backend metal layers - It is a primary structure for metal defect-density monitoring.…
metrology
Comb structure is an interdigitated test pattern for leakage detection — two comb-like fingers that approach without touching, creating high electric fields that accelerate…
Combining Multiple Anomaly Detection Methods Introduction & Motivation Combining Multiple Anomaly Detection Methods addresses a central problem in anomaly detection and {target}…
combinatorial test generation, combinatorial test, test generation
Combinatorial Test Generation Introduction & Motivation Combinatorial Test Generation addresses a central problem in verification process control and combination-count-explosion…
Combinatorics and Discrete Mathematics: Counting, Graphs, and Generating Functions Combinatorics develops systematic methods for counting, arranging, and selecting discrete…
reliability
Combined stress testing is application of multiple concurrent stresses such as temperature voltage and vibration during reliability testing - Concurrent stress exposure can reveal…
metrology
Combined Uncertainty ( ) is the total standard uncertainty of a measurement result obtained by combining all individual Type A and Type B uncertainty components — calculated using…
experiment, reproduce
Comet ML is the experiment tracking and model monitoring platform that provides deep hyperparameter optimization visualization and production model observability — enabling ML…
comet, evaluation
COMET is a learned evaluation metric that predicts translation quality using source hypothesis and reference representations - Neural regressors estimate human-like quality scores…
spin coating defect, photoresist particle streak, resist thickness non-uniformity, track module contamination, coat bowl particle
A comet defect is a streak of photoresist shaped exactly like its name — a bright head and a tapering tail trailing radially outward — left behind when a stray particle sitting on…
mlops
Comet.ml is the experiment tracking platform with strong emphasis on reproducibility, lineage capture, and comparative analytics - it automates collection of code, environment…
CSV, RFC 4180, TSV, CSV vs JSON vs Parquet, Excel gene name misinterpretation
Comma-separated values, or CSV, is a plain-text file format for tabular data in which each line represents a row and commas separate the values within that row, with the first…
shell command, command line, command-line arguments, options and flags, exit status, exit code, argv
A command is a single instruction typed into a shell or script that names a program, built-in function or alias to run, together with the extra information that tells it what…
CLI, command line, shell, teletype, Unix pipe, Bourne shell, command-line arguments
A command-line interface, or CLI, is a way of operating a computer by typing lines of text that name a program and its options, then reading the text it prints back. Instead…
code ai
Comments as Deodorant is a code smell where developers use comments to explain, justify, or apologize for code that is complex, unclear, or poorly structured — applying…
code ai
Commit Message Generation is the code AI task of automatically producing descriptive, informative git commit messages from code diffs — summarizing the semantic intent of source…
commit stage bottleneck and instruction retirement, commit stage bottleneck, stage bottleneck instruction, bottleneck instruction retirement, commit stage, stage bottleneck
Commit Stage Bottleneck and Instruction Retirement Introduction & Motivation Commit Stage Bottleneck and Instruction Retirement addresses a central problem in cpu-arch process…
commitment reveal protocol, commitment reveal, reveal protocol
Commitment Reveal Protocol Introduction & Motivation Commitment Reveal Protocol addresses a central problem in security-protocols process control and commitment-phase-binding…
reinforcement learning advanced
CommNet is a multi-agent architecture where agents communicate through differentiable shared message channels - Agent hidden states are aggregated and redistributed each step…
common-base amplifier, grounded-base, CB amplifier, cascode, current buffer, RF amplifier stage
A common-base amplifier is the bipolar-transistor stage in which the base is held at AC ground and acts as the terminal common to input and output: the signal is applied to…
spc
Common cause variation (also called random variation or natural variation) is the inherent, always-present variability in a process that results from the cumulative effect of many…
cross coupled layout, device matching layout, pelgrom mismatch model, dummy device guard ring, analog layout matching
Common-centroid layout is the answer to a problem no amount of careful transistor sizing can solve on its own: two devices meant to be identical are never placed at exactly…
common-collector amplifier, grounded-collector, CC amplifier, voltage follower configuration, emitter-follower configuration, buffer stage
A common-collector amplifier is the bipolar-transistor stage in which the collector terminal is held at AC ground — usually tied directly to the supply, which is a signal ground…
Common Crawl Foundation, CCBot, WARC, WET, WAT, GPT-3 training data, Gil Elbaz, C4, web corpus
Common Crawl is a nonprofit organization, founded in 2007 by Gil Elbaz, that crawls the public web and releases the results as a freely available, openly downloadable archive. It…
common-emitter amplifier, CE amplifier, voltage amplifier stage, phase inversion amplifier, voltage divider bias amplifier, grounded-emitter
A common-emitter amplifier is the bipolar-transistor stage in which the emitter terminal is common to the input and the output signal: the signal enters at the base, leaves at…
common-gate amplifier, common gate circuit, MOS common gate amplifier
Common-Gate Amplifier: Source Input, Drain Output A common-gate amplifier applies its input at a MOS transistor’s source, takes its output at the drain, and holds the gate near AC…
common-gate input impedance prediction, common-gate input impedance, input impedance prediction, common-gate input, input impedance, impedance prediction
Common-Gate Input Impedance Prediction with Machine Learning Introduction & Motivation Common-Gate Input Impedance Prediction with Machine Learning addresses a central problem…
common mode choke inductance drift prediction machine learning, common mode, choke inductance drift, emc emi, common mode choke inductance drift prediction, mode choke
Common Mode Choke Inductance Drift Prediction with Machine Learning Introduction & Motivation Common Mode Choke Inductance Drift Prediction with Machine Learning addresses…