Svcca
svcca, explainable ai
SVCCA is the representation comparison method combining singular value decomposition with canonical correlation analysis - it is used to compare learned subspaces between layers…
27,049 technical terms and definitions
svcca, explainable ai
SVCCA is the representation comparison method combining singular value decomposition with canonical correlation analysis - it is used to compare learned subspaces between layers…
svd++, recommendation systems
SVD++ is an extension of matrix factorization that incorporates implicit feedback into latent preference modeling - User factors are augmented with embeddings derived…
svd, model optimization
SVD Compression is a low-rank compression technique using singular value decomposition to truncate matrix components - It provides a principled way to retain dominant modes…
Singular Value Decomposition Recommendations Introduction & Motivation Singular Value Decomposition Recommendations addresses a central problem in recsys systems and {target}…
support vector, kernel
Support Vector Machine (SVM) is a supervised machine learning algorithm that finds the optimal hyperplane separating classes with the maximum margin — where the "support vectors"…
swag, optimization
SWAG (SWA-Gaussian) is an approximation to Bayesian deep learning that uses the SWA trajectory to fit a Gaussian distribution over weights — capturing both the mean (SWA solution)…
swag, evaluation
SWAG (Situations With Adversarial Generations) is the grounded commonsense inference benchmark — a 113,000-example dataset for predicting which of four sentence continuations is…
SWAG Stochastic Weight Averaging Gaussian Uncertainty Introduction & Motivation SWAG Stochastic Weight Averaging Gaussian Uncertainty addresses a central problem in uncertainty…
swap test for quantum state comparison, swap test quantum, test quantum state, quantum state comparison, swap test, test quantum, quantum state, state comparison
SWAP Test for Quantum State Comparison Introduction & Motivation SWAP Test for Quantum State Comparison addresses a central problem in quantum-algorithms process control…
swarm communication protocol, swarm communication, communication protocol
Swarm Communication Protocol Introduction & Motivation Swarm Communication Protocol addresses a central problem in robotics process control and message-passing-protocol-design…
swarm formation control, swarm formation, formation control
Swarm Formation Control Introduction & Motivation Swarm Formation Control addresses a central problem in robotics process control and formation-desired-shape-maintenance…
multi-agent
Swarm intelligence enables many simple agents to solve complex problems through emergent collective behavior. Inspiration: Ant colonies, bird flocks, bee hives - simple rules per…
swarm intelligence collective search and agent coordination, swarm intelligence collective, intelligence collective search, collective search agent, search agent coordination
Swarm Intelligence Collective Search and Agent Coordination Introduction & Motivation Swarm Intelligence Collective Search and Agent Coordination addresses a central problem…
swarm intelligence optimization, swarm intelligence, intelligence optimization
Swarm Intelligence Optimization Introduction & Motivation Swarm Intelligence Optimization addresses a central problem in ml process control and optimization-quality-convergence…
swarm path planning collective, swarm path planning, path planning collective, swarm path, path planning, planning collective
Swarm Path Planning Collective Introduction & Motivation Swarm Path Planning Collective addresses a central problem in robotics process control and collective-path-planning-group…
swarm robotics coordination, swarm robotics, robotics coordination
Swarm Robotics Coordination Introduction & Motivation Swarm Robotics Coordination addresses a central problem in swarm-robotics process control and swarm-size-scalability…
swarm robotics coordination, swarm robotics, robotics coordination
Swarm Robotics Coordination Introduction & Motivation Swarm Robotics Coordination addresses a central problem in robotics process control and multi-robot-coordination-collective…
self-supervised learning
SwAV (Swapping Assignments between Views) is a self-supervised learning method that combines contrastive learning with online clustering — assigning augmented views to prototype…
Swapped Assignment Clustering Introduction & Motivation Swapped Assignment Clustering addresses a central problem in selfsupervised and cluster-assignment-strategy architecture…
SwAV Unsupervised Learning of Visual Features by Online Clustering Introduction & Motivation SwAV Unsupervised Learning of Visual Features by Online Clustering addresses a central…
ai agents
SWE-bench is a benchmark for software-engineering agents that evaluates real bug-fix performance on code repositories - It is a core method in modern semiconductor AI-agent…
neural architecture
Activation functions are the reason depth means anything. Stack a hundred linear layers with no nonlinearity between them and the whole thing collapses algebraically into a single…
geglu activation, gated linear unit, ffn activation function, glu variant transformer
SwiGLU and GeGLU Activations are gated linear unit (GLU) variants that combine element-wise gating with smooth nonlinearities (Swish or GELU), achieving consistent improvements…
GLU variants, activation functions, transformer feed-forward, gating mechanism
SwiGLU and Gated Linear Units in Transformers are advanced activation architectures where feed-forward networks use gated mechanisms to selectively combine multiple transformation…
computer vision
Swin Transformer is the hierarchical vision transformer that makes self-attention practical for high-resolution images through shifted window attention — computing attention…
swin transformer hierarchical vision, swin transformer, transformer hierarchical, 4m ade20k coco fpn qk, swin transformer hierarchical, transformer hierarchical vision, hierarchical vision
Swin Transformer - Hierarchical Vision Introduction & Motivation Swin Transformer: hierarchical transformer with shifted windows. Local attention for efficiency. Applications…
Swin Transformer Hierarchical Vision Introduction & Motivation Swin Transformer Hierarchical Vision addresses a central problem in vision processing…
swin transformer shifted window and hierarchical attention prediction, swin transformer shifted, transformer shifted window, shifted window hierarchical, swin transformer
Swin Transformer Shifted Window and Hierarchical Attention Prediction Introduction & Motivation Swin Transformer Shifted Window and Hierarchical Attention Prediction addresses…
SimMIM Simple Masked Image Modeling on Swin Transformers Introduction & Motivation SimMIM Simple Masked Image Modeling on Swin Transformers addresses a central problem…
computer vision
SwinIR is the image restoration architecture based on Swin Transformer blocks for super-resolution, denoising, and artifact removal - it combines transformer context modeling…
multimodal ai
SwinIR is a transformer-based image restoration model for super-resolution, denoising, and artifact removal - It leverages shifted-window attention for efficient high-quality…
neural architecture
Activation functions are the reason depth means anything. Stack a hundred linear layers with no nonlinearity between them and the whole thing collapses algebraically into a single…
switch arbiter latency and grant pipeline depth, switch arbiter latency, arbiter latency grant, latency grant pipeline, grant pipeline depth, switch arbiter, arbiter latency, latency grant
Switch Arbiter Latency and Grant Pipeline Depth Introduction & Motivation Switch Arbiter Latency and Grant Pipeline Depth addresses a central problem in timing process control…
switch isolation insertion loss prediction, switch isolation insertion, isolation insertion loss, insertion loss prediction, switch isolation, isolation insertion, insertion loss
Switch Isolation Insertion Loss Prediction with Machine Learning Introduction & Motivation Switch Isolation Insertion Loss Prediction with Machine Learning addresses a central…
model architecture
Switch Transformer is a sparse Mixture of Experts (MoE) model architecture introduced by Fedus et al. (2022) at Google that simplifies MoE routing by sending each token to exactly…
architecture
Switch Transformer is mixture-of-experts transformer that routes each token to a single expert per sparse layer - It is a core method in modern semiconductor AI serving…
neural architecture
Switchable Normalization is a meta-normalization technique that learns to combine BatchNorm, InstanceNorm, and LayerNorm — using learnable weights to adaptively select the optimal…
switched-capacitor circuit noise and settling, switched-capacitor circuit noise, circuit noise settling, switched-capacitor circuit, circuit noise, noise settling
Switched-Capacitor Circuit Noise and Settling Introduction & Motivation Switched-Capacitor Circuit Noise and Settling addresses a central problem in analog process control…
switched-mode power supply, switched-mode power, power supply
Switched-Mode Power Supply Introduction & Motivation Switched-Mode Power Supply addresses a central problem in power-electronics process control and efficiency-percentage-load…
time series models
Switching State Space is state-space modeling with discrete regime switches and continuous within-regime dynamics. - It combines Markov switching logic with linear or nonlinear…
intel gpu sycl, sycl queue kernel, unified shared memory sycl, sycl interop cuda
SYCL and Intel oneAPI (DPC++): Standards-Based GPU Programming — cross-vendor portability and unified shared memory model SYCL is a Khronos-standardized C++17 abstraction layer…
GPU programming, unified, DPC++
SYCL oneAPI GPU Programming is a modern C++ framework enabling unified GPU programming across diverse hardware platforms through single-source compilation — supporting both…
sycl heterogeneous, oneapi cross platform, dpc++ programming
SYCL and oneAPI are modern programming frameworks for heterogeneous parallel computing that provide single-source C++ programming across CPUs, GPUs, FPGAs, and accelerators, using…
symbol timing recovery prediction, symbol timing recovery, timing recovery prediction, symbol timing, timing recovery, recovery prediction
Symbol Timing Recovery Prediction with Machine Learning Introduction & Motivation Symbol Timing Recovery Prediction with Machine Learning addresses a central problem in symbol…
symbolic aggregate approximation, symbolic aggregate, aggregate approximation
Symbolic Aggregate Approximation Introduction & Motivation Symbolic Aggregate Approximation addresses a central problem in timeseries process control…
symbolic ai knowledge graphs, symbolic ai, ai knowledge, ai, symbolic ai knowledge, ai knowledge graphs, knowledge graphs
Symbolic AI and Knowledge Graphs Introduction & Motivation Symbolic AI and Knowledge Graphs addresses a central problem in structured knowledge and reasoning: how to represent…
software engineering
Symbolic execution is a program analysis technique that executes programs with symbolic inputs rather than concrete values — exploring multiple execution paths simultaneously…
symbolic execution path explosion, symbolic execution path, execution path explosion, symbolic execution, execution path, path explosion
Symbolic Execution Path Explosion Introduction & Motivation Symbolic Execution Path Explosion addresses a central problem in verification process control…
symbolic fuzzing constraint solving, symbolic fuzzing constraint, fuzzing constraint solving, symbolic fuzzing, fuzzing constraint, constraint solving
Symbolic Fuzzing Constraint Solving Introduction & Motivation Symbolic Fuzzing Constraint Solving addresses a central problem in verification process control…
reasoning
Symbolic mathematics manipulates mathematical expressions as symbols rather than numeric values, enabling exact solutions, algebraic simplification, differentiation, integration…