awesome-approximate-dnn
github.com/e-dupuis/awesome-approximate-dnn ↗Curated content for DNN approximation, acceleration ... with a focus on hardware accelerator and deployment
Use this list with your AI agent
Add the Context Awesome MCP server to Claude, Cursor, or any MCP client, then ask:
"Show me pruning resources from awesome-approximate-dnn"
Installation instructions →What's inside
Approximation Methods
- Accelerating Sparse DNN Models without Hardware-Support via Tile-Wise SparsityPruning
Large matrix multiplication are tiled, this method propose to maintain a regular pattern at the tile level, improving efficiency.
- ALWANN: Automatic Layer-Wise Approximation of Deep Neural Network Accelerators without RetrainingApproximate operators
Use NSGA II to optimize approximate multipliers implemented & DNN mapping onto implemented Ax multipliers (Evo Approx).
- Automatic heterogeneous quantization of deep neural networks for low-latency inference on the edge for particle detectorsQuantization
AutoQKeras, Per layer quantization optimization using meta-heuristic DSE based on Bayesian Optimization, make use of Qkeras & hls4ml.
- Cross-Layer Approximation for Printed Machine Learning CircuitsMulti-techniques
Algorithmic and logic level approximation (coefficient replacement + netlist pruning) through a full DSE for printed ML applications.
- Deep Neural Network Compression by In-Parallel Pruning-QuantizationMulti-techniques
Use Bayesian optimization to solve both pruning and quantization problems jointly and with fine-tuning.
- Full Approximation of Deep Neural Networks through Efficient OptimizationApproximate operators
Select efficient approx multipliers through retraining and minimization of accuracy loss (Evo Approx)
Tools
- AdaptApproximations Frameworks
AdaPT is a fast emulation framework that extends PyTorch to support approximate inference as well as approximation-aware retraining
- BrevitasApproximations Frameworks
Pytorch extension to quantize DNN model
- codeDedicated Library
QNN inference library for ultra low power PULP RiscV core
- DistillerApproximations Frameworks
Distiller is an open-source Python package for neural network compression research (fine-tuning capable)
- DNN-NeurosimEvaluation Frameworks
Framework for evaluating the performance of inference or training of on-chip DNN
- DORYGraph Compiler
automatic tool to deploy DNNs on low-cost MCUs with typically less than 1MB of on-chip SRAM memory
Best Surveys
- Approximation Computing Techniques to Accelerate CNN Based Image Processing Applications – A Survey in Hardware/Software Perspective
- Deep Neural Network Approximation for Custom Hardware:Where We’ve Been, Where We’re Going
- Efficient Processing of Deep Neural Networks: A Tutorial and Survey
- Hardware Approximate Techniques for Deep Neural Network Accelerators: A Survey
- Model Compression and Hardware Acceleration for Neural Networks: A Comprehensive Survey
- Pruning and Quantization for Deep Neural Network Acceleration: A Survey
Others
- Blog postEfficient DNN Architecture
related to recent mobile architectures
- Google OR-ToolsGeneric DSE Framework
Constraint programming, routing and other optimization tools
- https://github.com/cedrickchee/awesome-ml-model-compressionSimilar repos
- https://github.com/chester256/Model-Compression-PapersSimilar repos
- https://github.com/guan-yuan/awesome-AutoML-and-Lightweight-ModelsSimilar repos
- https://github.com/he-y/Awesome-PruningSimilar repos
Showing a sample of 84 resources. View the full list on GitHub →