1,045 technical terms and definitions
gripper design finger, gripper design, design finger
Gripper Design Finger Introduction & Motivation Gripper Design Finger addresses a central problem in robotics process control and gripper-jaw-grasp-type prediction: how to Predict…
training phenomena
Grokking is the phenomenon where neural networks suddenly achieve perfect generalization on held-out data long after memorizing the training set and achieving near-zero training…
training phenomena
Grokking is a training phenomenon where a model suddenly generalizes long after memorizing the training data — the model first achieves perfect training accuracy (memorization)…
neural network grokking, double descent generalization, memorization to generalization transition, phase transition learning
Grokking and Delayed Generalization in Neural Networks is the phenomenon where a neural network first memorizes training data achieving perfect training accuracy, then much later…
cerebras, custom chip
Custom AI Accelerator Chips AI Chip Landscape Company Chip Focus NVIDIA H100, B200 General AI Groq LPU Low-latency inference Cerebras WSE-3 Largest chip, training Google TPU v5…
yield enhancement
Gross Die is the total number of potential die sites geometrically available on a wafer before yield loss - It defines theoretical output capacity at a given die size and wafer…
gross die per wafer forecast prediction, gross die per, die per wafer, per wafer forecast, wafer forecast prediction, gross die, die per, per wafer, wafer forecast, forecast prediction
Gross Die Per Wafer Forecast Prediction with Machine Learning Introduction & Motivation Gross Die Per Wafer Forecast Prediction with Machine Learning addresses a central problem…
industry
Gross margin is revenue minus cost of goods sold (COGS), expressed as a percentage of revenue. It measures how efficiently a semiconductor company converts revenue into profit…
business & strategy
Gross Margin is the percentage of revenue remaining after subtracting cost of goods sold, indicating core product profitability - It is a core method in advanced semiconductor…
design
Ground bounce (also called ground noise or simultaneous switching output noise on ground) is the transient voltage fluctuation on the ground (VSS) network caused by large, rapid…
signal & power integrity
Ground bounce is transient ground-potential variation caused by simultaneous switching currents through package and interconnect inductance - Rapid return-current changes create…
ground bounce ground plane design prediction, ground bounce ground, bounce ground plane, ground plane design, plane design prediction, ground bounce, bounce ground, ground plane
Ground Bounce Ground Plane Design Prediction with Machine Learning Introduction & Motivation Ground Bounce Ground Plane Design Prediction with Machine Learning addresses a central…
PCB ground plane, signal return plane, ground return path, ground stitching
Copper dual damascene interconnect architectures, electrochemical superfilling, and barrier-seed metallization constitute the back-end-of-line (BEOL) wiring systems that route…
design
Grounded-gate NMOS (GGNMOS) is the most widely used ESD protection clamp in CMOS technology, leveraging the parasitic lateral NPN bipolar transistor inherent in every NMOS device…
rag
Grounded generation is the response generation approach that constrains model output to provided evidence rather than unconstrained parametric memory - it is a primary method…
robotics
Grounded Language Learning is the AI research paradigm that acquires language understanding through interaction with physical or simulated environments — learning word…
evaluation
Groundedness is the extent to which generated claims are supported by provided context or verifiable external sources - It is a core method in modern AI fairness and evaluation…
factual, knowledge
Grounding LLM Responses What is Grounding? Grounding ensures LLM outputs are based on reliable sources rather than model parameters alone. It bridges the gap between fluent…
rag
Grounding ensures AI outputs are anchored in retrieved facts rather than generated from potentially unreliable model knowledge. Problem: LLMs may generate plausible but false…
manufacturing operations
Grounding is the creation of low-impedance electrical paths that safely drain static charge to earth reference - It is a core method in modern semiconductor wafer handling…
facility
Grounding and bonding is the electrical interconnection of all conductive objects within an ESD Protected Area to a common earth ground reference — ensuring that no metal fixture…
computer vision
Grounding DINO is a state-of-the-art open-set object detector — combining the transformer-based detection of DINO (DETR variant) with grounded pre-training to detect arbitrary…
rag
Grounding in external knowledge is the practice of anchoring responses in retrieved evidence rather than relying only on model memory - Retrieval pipelines fetch supporting…
neural architecture
Group Convolutions (G-Convolutions) are the mathematical generalization of standard convolution from the translation group to arbitrary symmetry groups — including rotation…
group delay frequency dispersion prediction, group delay frequency, delay frequency dispersion, frequency dispersion prediction, group delay, delay frequency
Group Delay Frequency Dispersion Prediction with Machine Learning Introduction & Motivation Group Delay Frequency Dispersion Prediction with Machine Learning addresses a central…
Group Normalization Variants Introduction & Motivation Group Normalization Variants addresses a central problem in vision processing and small-batch-stability prediction: how…
recommendation systems
Group Recommendation is recommendation for multi-user groups instead of single-user personalization - It aggregates member preferences to rank items acceptable to the group as…
Group Recommendation Systems Introduction & Motivation Group Recommendation Systems addresses a central problem in recsys systems and {target} optimization: how to Build {subject}…
leak, prevent
GroupKFold is a cross-validation strategy that prevents data leakage by ensuring all samples from the same "group" stay together in either the training set or the test set, never…
computer vision
Grouped Convolution is a convolution where input channels are divided into groups, and each group is convolved independently — reducing parameters and FLOPs by a factor of while…
model optimization
Grouped Convolution is a convolution method that partitions channels into groups processed by separate filter sets - It reduces parameters and compute while preserving…
gqa, kv
Multi-head attention (MHA), multi-query attention (MQA), and grouped-query attention (GQA) are three ways to wire the key and value projections of a Transformer's attention layer.…
gqa, multi query attention, mqa, attention head sharing
Grouped-Query Attention (GQA) is the attention architecture variant that shares Key and Value heads among groups of Query heads — reducing the KV cache memory footprint…
grouped-query attention, gqa, llm architecture
Grouped-Query Attention (GQA) is an attention architecture that provides a tunable middle ground between Multi-Head Attention (MHA) and Multi-Query Attention (MQA) — using G…
multi query attention mqa, kv cache reduction, attention head grouping, llama 2 attention
Grouped Query Attention (GQA) is the attention mechanism that shares key and value projections across groups of query heads, interpolating between multi-head attention (MHA)…
optimization
Grouped-query KV cache is the attention approach where query heads are partitioned into groups that share key-value heads, balancing efficiency between full multi-head attention…
split-apply-combine, Hadley Wickham, GROUP BY, pivot table, window functions
Grouping is the operation of splitting a dataset into subsets that share a common value in one or more columns, so that a calculation can be applied separately to each subset…
neural architecture
GroupNorm is a normalization technique that divides channels into groups and normalizes within each group — independent of batch size, making it the preferred normalization…
quantum ai
Grover's Algorithm is a quantum search algorithm that finds a marked item in an unsorted database of N elements using only O(√N) queries to the database oracle, achieving…
grover search algorithm, grover search, search algorithm
Grover Search Algorithm Introduction & Motivation Grover Search Algorithm addresses a central problem in quantum process control and search-algorithm-quadratic-speedup prediction…
grover's search and optimization, grover's search optimization, grover's search, search optimization
Grover's Search and Optimization Introduction & Motivation Grover's Search and Optimization addresses a central problem in quantum-algorithms process control…
grown junction 1951 first n type region, quiet growth interval czochralski, emitter collector region length, grown junction transistor outer region growth
Grow the First N-Type Region: The First Quiet Interval in an Alternating Process Nothing chemically new happens during this step — the melt is still the same n-type composition…
grown junction 1951 second n type region, collector depletion width reverse bias, reach through breakdown transistor, grown junction transistor collector length
Grow the Second N-Type Region: A Quiet Interval With a Voltage-Dependent Constraint Step twelve's quiet interval had no electrical constraint at all; this one has a constraint…
grown junction 1951 thin p type base, base width current gain exponential, grown junction transistor gain sensitivity, diffusion length base width ratio
Grow the Thin P-Type Base: The Quiet Interval That Isn't Allowed to Drift Step twelve's quiet interval had no electrical constraint at all — this one has the tightest constraint…
grown junction transistor, Sparks and Teal 1950, doped crystal pulling, melt doping, n-p-n germanium transistor, Shockley Sparks Teal 1951 paper
A grown-junction transistor is a bipolar junction transistor whose p-n junctions are formed inside a single crystal while it is being pulled from a molten semiconductor…
grown-junction transistor process flow, grown junction transistor device process flow, grown junction transistor process, Czochralski melt dosed pull, Teal Sparks grown junction, pull rate base width control
Grown-Junction Transistor Device Process Flow: Seed-Dip Timing, Pull-Rate Kinetics & Dopant-Switch Yield Control The grown-junction transistor was the first transistor…
rpc, streaming
gRPC is the high-performance Remote Procedure Call framework developed by Google that uses HTTP/2 for transport and Protocol Buffers for serialization — enabling efficient…
group relative policy optimization, llm reward free rl, process reward model training, math reasoning rl
GRPO and RL for LLM Reasoning is the reinforcement learning training paradigm that directly optimizes large language models for verifiable reasoning tasks — particularly…
gated recurrent unit, Cho 2014, update gate, reset gate, GRU vs LSTM, gated rnn, recurrent neural network
A gated recurrent unit, or GRU, is a recurrent neural network cell that reads a sequence one step at a time while carrying a hidden state, and that uses two learned gates…
gru gated recurrent, gru gated, gated recurrent
GRU Gated Recurrent Introduction & Motivation GRU Gated Recurrent addresses a central problem in timeseries process control and gru-reset-update-gate prediction: how to Predict…