MIMUW AI research lab's repositories
Unified-Long-Horizon-Time-Series-Benchmark
Code for reproducing results from the paper "Unified Long Horizon Time Series Benchmark"
FreTS-fork
Official implementation of the paper "Frequency-domain MLPs are More Effective Learners in Time Series Forecasting"
improved_overparametrization_tmlr
Code for reproducing experiments reported in the TMLR paper 'Improved Overparametrization Bounds for Global Convergence of Stochastic Gradient Descent for Shallow Neural Networks'
marl_team_based_testers
Python scripts dedicated for testing team-based multi-agent reinforcement learning environments. Behavioral policies of agents are provided as ONNX checkpoints. Currently compatible with Unity ML-Agents framework.
ml-agents_fork
The Unity Machine Learning Agents Toolkit (ML-Agents) is an open-source project that enables games and simulations to serve as environments for training intelligent agents using deep reinforcement learning and imitation learning.
unity-ml-agents_hide-and-seek
New Unity ML-Agents Team-based MARL environments: Hide & Seek and Predator-Prey. Used in the paper "FCSP: Fictitious Co-Self-Play for Team-based Multi-agent Reinforcement Learning"
unity_dodgeball_training_fork
Repository with environment C# code for training GFCP algorithm (team self-play with checkpoint injections) using ML-Agents learn script
worrisome-nn
Code repository for reproducing results published in ECAI23 paper entitled "Worrisome Properties of Neural Network Controllers and Their Symbolic Representations"