Links

Efficient and Sustainable Machine Learning:

• Algorithms for efficient deep learning (Google Research, 2022 & beyond).
• AutoML: Automation of machine learning workshop.
• BEARD: BEnchmarking the Adversarial Robustness for Dataset distillation.
• DC-Bench: A standardized benchmark for dataset condensation.
• In AI, is bigger always better? (Nature).
• MIT HAN Lab: Efficient AI computing lab at MIT.
• Numenta: A young company about neuroscience and AI.

Seminars on Optimization:

• BrainPOP: brainstorming days on measure and polynomial optimization.
• OWOS: One World Optimization Seminar.
• POEMA: Polynomial Optimization, Efficiency through Moments and Algebra.
• SPOT: Séminaire Pluridisciplinaire d’Optimisation de Toulouse.

Polynomial Optimization Software:

• Gloptipoly: generalized problem of moments (GPM) solver.
• Gurobi: mathematical programming solver.
• Mosek: large scale optimization software.
• STRIDE: SpecTrahedRal Inexact projected gradient Descent along vErtices.
• TSSOS: polynomial optimization tool based on moment-SOS hierarchy.
• Yalmip: a toolbox for modeling and optimization in MATLAB.

Robustness Verification Software:

• alpha, beta-CROWN: A fast and scalable neural network verifier with efficient bound propagation.
• AutoAttack: State-of-the-art automated adversarial attacks.
• convex_adversarial: Training provably robust neural networks by optimizing convex outer bounds on the adversarial polytope.
• ERAN: ETH Robustness Analyzer for Neural networks.
• Foolbox: Fast adversarial attacks to benchmark the robustness of machine learning models by PyTorch, TensorFlow, and JAX.
• jax_verify: Neural network verification in JAX.
• MultiRobustBench: A standardized benchmark for evaluating adversarial robustness against multiple attacks.
• RobustBench: A standardized benchmark for adversarial robustness using AutoAttack (which is the starting point of RobustBench and is no longer maintained.
• RobustML: A community-run hub for learning about robust machine learning.
• RobustNeuralNetworks.jl: A Julia package for robust neural networks built from the Recurrent Equilibrium Network (REN) and Lipschitz-Bounded Deep Network (LBDN) model classes.
• SoK: Certified robustness for deep neural networks.

Workshops and Tutorials on Verification:

• Adversarial Robustness - Theory and Practice by Zico Kolter and Aleksander Madry.
• Reliable and Interpretable Artificial Intelligence by Martin Vechev.
• Trustworthy Machine Learning by Reza Shokri and Nicolas Papernot.
• VNN-COMP: International Verification of Neural Networks COMPetition.
• WFVML: Workshop on Formal Verification of Machine Learning.