Computational Efficiency
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
- 하위 태스크
- 0
- 평가표
- 1
- 보존된 평가 기록
- 1
Methods and optimizations to reduce the computational resources (e.g., time, memory, or power) needed for training and inference in models. This involves techniques that streamline processing, optimize algorithms, or leverage hardware to enhance performance without compromising accuracy.
코드
관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.
https://github.com/dpiras/cosmopower-jax
공식 코드CosmoPower-JAX: high-dimensional Bayesian inference with differentiable cosmological emulators
https://github.com/dqzhang17/clcnet.torch
공식 코드clcNet: Improving the Efficiency of Convolutional Neural Network using Channel Local Convolutions
https://github.com/dragonlee258079/DMT
공식 코드Discriminative Co-Saliency and Background Mining Transformer for Co-Salient Object Detection
https://github.com/dreamy-xay/tablecenternet
공식 코드TableCenterNet: A one-stage network for table structure recognition
https://github.com/drosemei/rome
공식 코드RoMe: Towards Large Scale Road Surface Reconstruction via Mesh Representation
https://github.com/drscotthawley/signaltrain
공식 코드Exploring Quality and Generalizability in Parameterized Neural Audio Effects
https://github.com/ds3lab/corgipile-postgresql
공식 코드https://github.com/ds4dh/GLiNER-biomed
공식 코드GLiNER-BioMed: A Suite of Efficient Models for Open Biomedical Named Entity Recognition
https://github.com/dslaborg/germeval2024
공식 코드https://github.com/dslisleedh/ESC
공식 코드Emulating Self-attention with Convolution for Efficient Image Super-Resolution
https://github.com/dslisleedh/IGConv
공식 코드Implicit Grid Convolution for Multi-Scale Image Super-Resolution
https://github.com/dslisleedh/PLKSR
공식 코드https://github.com/duanlab123/mgmc
공식 코드https://github.com/duo-yu/neural-gode
공식 코드Neural Generalized Ordinary Differential Equations with Layer-varying Parameters
https://github.com/dyahayumgw/HAAQI-Net
공식 코드HAAQI-Net: A Non-intrusive Neural Music Audio Quality Assessment Model for Hearing Aids
https://github.com/dyu62/PolyGNN
공식 코드PolygonGNN: Representation Learning for Polygonal Geometries with Heterogeneous Visibility Graph
https://github.com/eBay/maxDNN
공식 코드maxDNN: An Efficient Convolution Kernel for Deep Learning with Maxwell GPUs
https://github.com/eashanadhikarla/expomamba
공식 코드ExpoMamba: Exploiting Frequency SSM Blocks for Efficient and Effective Image Enhancement
https://github.com/eatbeanss/CSWin-UNet
공식 코드CSWin-UNet: Transformer UNet with Cross-Shaped Windows for Medical Image Segmentation
https://github.com/ebekkers/ponita
공식 코드Fast, Expressive SE$(n)$ Equivariant Networks through Weight-Sharing in Position-Orientation Space
https://github.com/ebilionis/variational-reformulation-of-inverse-problems
공식 코드https://github.com/ecnuacrush/hyatt-net
공식 코드HYATT-Net is Grand: A Hybrid Attention Network for Performant Anatomical Landmark Detection
https://github.com/ed-fish/Geo-Sign
공식 코드Geo-Sign: Hyperbolic Contrastive Regularisation for Geometrically Aware Sign Language Translation
https://github.com/efm95/dream
공식 코드Modeling non-linear Effects with Neural Networks in Relational Event Models
https://github.com/ehsanhaghighat/stonet
공식 코드STONet: A novel neural operator for modeling solute transport in micro-cracked reservoirs
https://github.com/eidoslab/towards-efficient-capsule-networks
공식 코드https://github.com/eitanspi/weather-image-classification
공식 코드https://github.com/ejhigson/bsr
공식 코드Bayesian sparse reconstruction: a brute-force approach to astronomical imaging and machine learning
https://github.com/ejokhan/transnas_tsad
공식 코드https://github.com/ejwww/tangnn
공식 코드https://github.com/elasalle/pasco
공식 코드PASCO (PArallel Structured COarsening): an overlay to speed up graph clustering algorithms
https://github.com/eleurent/highway-env
공식 코드https://github.com/elfsong/mercury
공식 코드Mercury: A Code Efficiency Benchmark for Code Large Language Models
https://github.com/elkabzo/cont_disc_opt_dnn
공식 코드Continuous vs. Discrete Optimization of Deep Neural Networks
https://github.com/elki-project/elki
공식 코드출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.