15,414 technical terms and definitions
dataflow computing and architectures, dataflow computing architectures, dataflow computing, computing architectures
Dataflow Computing and Architectures Introduction & Motivation Dataflow Computing and Architectures addresses a central problem in dataflow-computing process control…
wave computing, spatial architecture computing, coarse grain reconfigurable array cgra, stream dataflow architecture
Dataflow Processor Architecture: Spatial Computing via Coarse-Grained Reconfigurable Arrays — compute elements directly mapped to hardware nodes with data-driven execution model…
corpus, training data
Training Data for LLMs Pretraining Datasets Large language models are pretrained on massive text corpora—often trillions of tokens from diverse sources. Common Pretraining Sources…
Dataset Cartography and Data Maps for Training Dynamics Introduction & Motivation Dataset Cartography and Data Maps for Training Dynamics addresses a central problem…
distributed training
Dataset sharding is the partitioning of training data into non-overlapping subsets assigned across distributed workers - it ensures balanced workload distribution, minimizes…
Date-Time Parsing and Normalization Introduction & Motivation Date-Time Parsing and Normalization addresses a central problem in {subject} assurance and {target} management: how…
d2d variation, daily drift
Day-to-Day Variation (D2D) in semiconductor manufacturing refers to process parameter fluctuations between production days caused by environmental, equipment, or operational…
dbr mirror layer thickness drift prediction, dbr mirror layer, mirror layer thickness, layer thickness drift, thickness drift prediction, dbr mirror, mirror layer
DBR Mirror Layer Thickness Drift Prediction with Machine Learning Introduction & Motivation DBR Mirror Layer Thickness Drift Prediction with Machine Learning addresses a central…
dbr mirror stack uniformity prediction machine learning, dbr mirror, stack uniformity prediction, dbr vcsel, dbr mirror stack uniformity prediction, dbr mirror stack
DBR Mirror Stack Uniformity Prediction with Machine Learning Introduction & Motivation DBR Mirror Stack Uniformity Prediction with Machine Learning addresses a central problem…
dbscan clustering anomaly, dbscan clustering, clustering anomaly
DBSCAN Clustering Anomaly Introduction & Motivation DBSCAN Clustering Anomaly addresses a central problem in timeseries process control and density-based-clustering-method…
data center hardware, data center, center hardware
Data Center Hardware Introduction & Motivation Data Center Hardware addresses a central problem in data-center process control and pue prediction: how to Predict pue from dc-hw…
ddim, denoising diffusion implicit models, generative models
DDIM (Denoising Diffusion Implicit Models) is an accelerated sampling method for diffusion models that defines a family of non-Markovian diffusion processes sharing the same…
ddim, generative models
DDIM sampling is the non-Markov diffusion sampling method that enables deterministic or partially stochastic generation with fewer steps - it reuses DDPM-trained models while…
drift transistor 1956 wafer cleaning lapping cmp, double diffused mesa transistor 1956 contamination trap leakage, silicon surface preparation diffusion readiness drift transistor
Surface Clean and Polish the Wafer: Removing Everything a Diffusion Furnace Would Otherwise Copy Into the Crystal 1. Why a Furnace Cannot Be Trusted to Ignore a Dirty Surface…
drift transistor 1956 sheet resistance junction depth check, double diffused mesa transistor 1956 base profile verification, drift field measurement before emitter diffusion silicon mesa
Verify the Base Drift-Field Profile: Measuring a Gradient Before Trusting It With an Emitter 1. Why a Field You Cannot Measure Is a Field You Cannot Rely On Before any emitter…
drift transistor 1956 emitter shallow heavy doping, double diffused mesa transistor 1956 emitter base junction placement, phosphorus emitter diffusion drift field silicon mesa 1956
Emitter Predeposition: A Second Dopant, Opposite Polarity, Placed Inside a Field That Is Already Waiting For It 1. Why the Emitter Must Be Shallower and More Heavily Doped Than…
ddos mitigation strategies, ddos mitigation, mitigation strategies
DDoS Mitigation Strategies Introduction & Motivation DDoS Mitigation Strategies addresses a central problem in network-security process control and mitigation-latency-seconds…
ddos protection mitigation, ddos protection, protection mitigation
DDoS Protection Mitigation Introduction & Motivation DDoS Protection Mitigation addresses a central problem in cloud process control and attack-detection-latency prediction: how…
dielectric deposition, high-k dielectrics, ald, pecvd, gap fill, hdpcvd, feature-scale modeling
Semiconductor Manufacturing: Dielectric Deposition Process (DDP) Modeling Overview DDP (Dielectric Deposition Process) refers to the set of techniques used to deposit insulating…
ddpm, generative models
DDPM is the Denoising Diffusion Probabilistic Model framework that learns a reverse Markov chain from noisy data to clean samples - it established the modern baseline…
ddr controller refresh command accuracy prediction, ddr controller refresh, controller refresh command, refresh command accuracy, command accuracy prediction
DDR Controller Refresh Command Accuracy Prediction with Machine Learning Introduction & Motivation DDR Controller Refresh Command Accuracy Prediction with Machine Learning…
ddr interface pre-emphasis tuning prediction, ddr interface pre-emphasis, interface pre-emphasis tuning, pre-emphasis tuning prediction, ddr interface, pre-emphasis tuning
DDR Interface Pre-Emphasis Tuning Prediction with Machine Learning Introduction & Motivation DDR Interface Pre-Emphasis Tuning Prediction with Machine Learning addresses a central…
ddr memory interface timing margin prediction, ddr memory interface, memory interface timing, interface timing margin, timing margin prediction, ddr memory, memory interface
DDR Memory Interface Timing Margin Prediction with Machine Learning Introduction & Motivation DDR Memory Interface Timing Margin Prediction with Machine Learning addresses…
ddr5 memory, ddr5 dimm, dram interface, ddr5 training
DDR5 is the fifth generation of double-data-rate synchronous DRAM, designed to increase bandwidth, density, channel efficiency, and reliability for servers and client systems.…
ddr5 timing margin prediction, ddr5 timing margin, timing margin prediction, ddr5 timing, timing margin, margin prediction
DDR5 Timing Margin Prediction with Machine Learning Introduction & Motivation DDR5 Timing Margin Prediction with Machine Learning addresses a central problem in DDR5 memory timing…
healthcare ai
De Novo Drug Design is the generative AI approach to creating entirely new drug molecules from scratch — molecules that do not exist in any database — optimized to satisfy…
model optimization
Dead Code Elimination is removing graph nodes and branches that do not affect final outputs - It streamlines execution graphs and reduces unnecessary compute. What Is Dead Code…
dead code elimination, dead code, code elimination
Dead Code Elimination Introduction & Motivation Dead Code Elimination addresses a central problem in compiler-optimization process control and elimination-coverage prediction: how…
deadline-driven scheduling algorithms, deadline-driven scheduling, scheduling algorithms
Deadline-Driven Scheduling Algorithms Introduction & Motivation Deadline-Driven Scheduling Algorithms addresses a central problem in real-time-systems process control…
ai safety
Debate is an AI alignment approach where two AI agents argue opposing sides of a question, and a human judge selects the most compelling argument — the key insight is that even if…
ai safety
Debate is an alignment protocol where competing AI agents argue opposing claims for a judge to evaluate - It is a core method in modern AI safety execution workflows. What Is…
foundation model
DeBERTa (Decoding-enhanced BERT with Disentangled Attention) is a pre-trained language model that improves upon BERT by disentangling content and position representations…
Debiased Pseudo-Labeling for Long-Tailed Semi-Supervised Introduction & Motivation Debiased Pseudo-Labeling for Long-Tailed Semi-Supervised addresses a central problem…
ai safety
Debiasing Techniques are methods for reducing or eliminating unwanted biases in AI systems across the machine learning pipeline — encompassing pre-processing approaches that…
fairness
Debiasing techniques is the algorithmic and data-centric methods used to reduce biased associations in model representations and outputs - debiasing targets both learned internal…
debug interface system integration and observability, debug interface system, interface system integration, system integration observability, debug interface, interface system
Debug Interface System Integration and Observability Introduction & Motivation Debug Interface System Integration and Observability addresses a central problem in debug process…
debug probe insertion, debug probe, probe insertion
Debug Probe Insertion Introduction & Motivation Debug Probe Insertion addresses a central problem in verification process control and probe-insertion-latency-seconds prediction…
debug turnaround time and efficiency, debug turnaround time, turnaround time efficiency, debug turnaround, turnaround time, time efficiency
Debug Turnaround Time and Efficiency Introduction & Motivation Debug Turnaround Time and Efficiency addresses a central problem in debug process control and debug-cycle-time-hours…
troubleshooting, eval sets, logging, tracing, langsmith, prompt engineering
Debugging LLM applications is the systematic process of identifying and fixing issues in AI-powered systems — addressing problems like hallucinations, format errors, inconsistent…
debugging tool effectiveness, debugging tool, tool effectiveness
Debugging Tool Effectiveness Introduction & Motivation Debugging Tool Effectiveness addresses a central problem in programming process control and debugger-overhead-performance…
Decentralized Federated Learning Introduction & Motivation Decentralized Federated Learning addresses a central problem in federated and peer-to-peer-aggregation architecture: how…
decentralized optimization gossip prediction, decentralized optimization gossip, optimization gossip prediction, decentralized optimization, optimization gossip, gossip prediction
Decentralized Optimization Gossip Prediction with Machine Learning Introduction & Motivation Decentralized Optimization Gossip Prediction with Machine Learning addresses a central…
Peer-to-Peer Gossip Learning Without Central Server Introduction & Motivation Peer-to-Peer Gossip Learning Without Central Server addresses a central problem in federated…
Decentralized POMDP Planning under Communication Bounds Introduction & Motivation Decentralized POMDP Planning under Communication Bounds addresses a central problem…
decimation filter aliasing margin prediction, decimation filter aliasing, filter aliasing margin, aliasing margin prediction, decimation filter, filter aliasing
Decimation Filter Aliasing Margin Prediction with Machine Learning Introduction & Motivation Decimation Filter Aliasing Margin Prediction with Machine Learning addresses a central…
Decision Boundary Visualization Introduction & Motivation Decision Boundary Visualization addresses a central problem in explainability interpretation and model understanding: how…
decision making under uncertainty, decision making under, making under uncertainty, decision making, making under, under uncertainty
Decision Making Under Uncertainty Introduction & Motivation Decision Making Under Uncertainty addresses a central problem in autonomous-vehicle process control…
decision table testing, decision table, table testing
Decision Table Testing Introduction & Motivation Decision Table Testing addresses a central problem in verification process control and table-completeness-percentage prediction…
Decision Transformer for Trajectory Sequence Modeling Introduction & Motivation Decision Transformer for Trajectory Sequence Modeling addresses a central problem in reinforcement…
explainable ai
Decision Tree Extraction is a model distillation technique that trains a decision tree to approximate the predictions of a complex model — producing an interpretable…