adversarial-robustness-toolbox
alpha-zero-boosted
Our great sponsors
adversarial-robustness-toolbox | alpha-zero-boosted | |
---|---|---|
8 | 2 | |
4,433 | 79 | |
2.3% | - | |
9.7 | 3.2 | |
5 days ago | over 3 years ago | |
Python | Python | |
MIT License | - |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
adversarial-robustness-toolbox
- [D] Couldn't devs of major GPTs have added an invisible but detectable watermark in the models?
- [D] ML Researchers/Engineers in Industry: Why don't companies use open source models more often?
- [D]: How safe is it to just use a strangers Model?
-
[D] Does anyone care about adversarial attacks anymore?
Check out this project https://github.com/Trusted-AI/adversarial-robustness-toolbox
- adversarial-robustness-toolbox: Adversarial Robustness Toolbox (ART) - Python Library for Machine Learning Security - Evasion, Poisoning, Extraction, Inference - Red and Blue Teams
- Library for Machine Learning Security Evasion, Poisoning, Extraction, Inference
-
Introduction to Adversarial Machine Learning
Adversarial Robustness Toolbox (ART) is a Python library for Machine Learning Security. ART provides tools that enable developers and researchers to defend and evaluate Machine Learning models and applications against the adversarial threats of Evasion, Poisoning, Extraction, and Inference.
-
[D] Testing a model's robustness to adversarial attacks
Depending on what attacks you want I've found both https://github.com/cleverhans-lab/cleverhans and https://github.com/Trusted-AI/adversarial-robustness-toolbox to be useful.
alpha-zero-boosted
-
DeepMind has open-sourced the heart of AlphaGo and AlphaZero
> I came up with a nifty implementation in Python that outperforms the naive impl by 30x, allowing a pure python MCTS/NN interop implementation. See https://www.moderndescartes.com/essays/deep_dive_mcts/
Great post!
Chasing pointers in the MCTS tree is definitely a slow approach. Although typically there are < 900 "considerations" per move for alphazero. I've found getting value/policy predictions from a neural network (or GBDT[1]) for the node expansions during those considerations is at least an order of magnitude slower than the MCTS tree-hopping logic.
-
MuZero: Mastering Go, chess, shogi and Atari without rules
What you can do is checkout the algorithm at a particular stages of development. AlphaZero&Friends start out not being very good at the game, then over time they learn and become super human. You typically checkpoint the weights for the model at various stages. So early on, the algo would be like a 600 elo player for chess and then eventually get to superhuman elo levels. So if you wanted to train you can gradually play against versions of the algo until you can beat them by loading up the weights at various difficulty stages.
I implemented AlphaZero (but not Mu yet) using GBDTs instead of NNs here if you're curious about how it would work: https://github.com/cgreer/alpha-zero-boosted. Instead of saving the "weights" for a GBDT, you save the splitpoints for the value/policy models, but the concept is the same.
What are some alternatives?
DeepRobust - A pytorch adversarial library for attack and defense methods on images and graphs
KataGo - GTP engine and self-play learning in Go
auto-attack - Code relative to "Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attacks"
neural_network_chess - Free Book about Deep-Learning approaches for Chess (like AlphaZero, Leela Chess Zero and Stockfish NNUE)
TextAttack - TextAttack 🐙 is a Python framework for adversarial attacks, data augmentation, and model training in NLP https://textattack.readthedocs.io/en/master/
katrain - Improve your Baduk skills by training with KataGo!
m2cgen - Transform ML models into a native code (Java, C, Python, Go, JavaScript, Visual Basic, C#, R, PowerShell, PHP, Dart, Haskell, Ruby, F#, Rust) with zero dependencies
leela-zero - Go engine with no human-provided knowledge, modeled after the AlphaGo Zero paper.
waf-bypass - Check your WAF before an attacker does
mars - Mars is a tensor-based unified framework for large-scale data computation which scales numpy, pandas, scikit-learn and Python functions.
Differential-Privacy-Guide - Differential Privacy Guide
mctx - Monte Carlo tree search in JAX