Name: Grigori Fursin
Type: User
Company: Flex.ai, cTuning.org and MLCommons
Bio: Head of FlexAI R&D Lab co-designing more efficient, cost-effective and reliable software and hardware for AI, ML and other emerging workload
Twitter: grigori_fursin
Location: Paris
Blog: https://cKnowledge.org/gfursin
Grigori Fursin's Projects
Byzantine-resilient distributed SGD with TensorFlow.
An awesome list of high-quality open datasets in public domains (on-going).
Browser extension for Chrome and FireFox to collaboratively annotate web pages with related research papers, code, reproducible results, scoreboards, portable workflows and reusable artifacts.
Caffe2 is a lightweight, modular, and scalable deep learning framework.
Collective Knowledge DataBase
Collective Knowledge framework (CK) helps to organize any software project as a database of reusable components and micro-services with common APIs, automation actions and extensible meta descriptions. See real-world use cases from MLPerf, Arm, General Motors, IBM, Raspberry Pi and ACM:
Collective Knowledge extension to let users implement customizable, portable, multi-dimensional and multi-objective SW/HW auto-tuning workflows using Collective Knowledge Framework. It is used as a universal engine for collaborative AI crowd-benchmarking and crowd-tuning across different compilers, libraries, run-time systems and platforms:
CK dissemination
CK-NNTest: collaboratively validating, benchmarking and optimizing neural net operators across platforms, frameworks and datasets
Collective Knowledge workflows for OpenVINO Toolkit (Deep Learning Deployment Toolkit)
Data for portal
CK workflow for ReQuEST ASPLOS'18 submission:
Collective Mind Repository (cTuning; Benchmarks) - outdated
Collective Mind Repository (cTuning; Program data sets) - outdated
Collective Mind Repository (cTuning; Main) - outdated
http://ctuning.org/cm/wiki/index.php?title=CM:data:45741e3fbcf4024b:1db78910464c9d05
http://ctuning.org/cm/wiki/index.php?title=CM:data:45741e3fbcf4024b:1db78910464c9d05
http://ctuning.org/cm/wiki/index.php?title=CM:data:45741e3fbcf4024b:1db78910464c9d05
A reinforcement learning toolkit for compiler optimizations
Prototyping connectME library
Deep Learning Specialization by Andrew Ng on Coursera.
dividiti's OpenCL API profiler
A toolkit for developing and comparing reinforcement learning algorithms.
Reference implementations of MLPerf™ inference benchmarks
MLPerf™ v1.0 results
Empty repository for MLPerf inference results
API Client for paperswithcode.com
A pytorch Quantization Toolkit
Tips for releasing research code in Machine Learning (with official NeurIPS 2020 recommendations)