MarcyLee

gitignore

A collection of useful .gitignore templates

ray

Ray is a unified framework for scaling AI and Python applications. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

tianshou

An elegant PyTorch deep reinforcement learning library.

sacred

Sacred is a tool to help you configure, organize, log and reproduce experiments developed at IDSIA.

modAL

A modular active learning framework for Python

pymarl

Python Multi-Agent Reinforcement Learning framework

maddpg

Code for the MADDPG algorithm from the paper "Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments"

continual-learning

PyTorch implementation of various methods for continual learning (XdG, EWC, SI, LwF, FROMP, DGR, BI-R, ER, A-GEM, iCaRL, Generative Classifier) in three different scenarios.

Ipopt

COIN-OR Interior Point Optimizer IPOPT

active-learning

deep-active-learning

Deep Active Learning

simple_canvas_game

Quick tutorial on how to make a simple HTML5 Canvas game

pytorch-trpo

PyTorch implementation of Trust Region Policy Optimization

mbbl

softgym

SoftGym is a set of benchmark environments for deformable object manipulation.

marLo

Multi Agent Reinforcement Learning using MalmÖ

ewc.pytorch

An implementation of EWC with PyTorch

urdf_tutorial

multiagent-gail

robovat

RoboVat: A unified toolkit for simulated and real-world robotic task environments.

Multi-agent-reinforcement-learning

Implementation of Multi-Agent Reinforcement Learning algorithm(s). Currently includes: MADDPG

tianshou-docs-zh_CN

天授中文文档

Dynamic-Movement-Primitives-and-Imitation-Learning-Robotics

Dynamic movement primitives (DMPs) are a method of trajectory control/planning from Stefan Schaal’s lab. Complex movements have long been thought to be composed of sets of primitive action ‘building blocks’ executed in sequence and \ or in parallel, and DMPs are a proposed mathematical formalization of these primitives. The difference between DMPs and previously proposed building blocks is that each DMP is a nonlinear dynamical system. The basic idea is that you take a dynamical system with well specified, stable behavior and add another term that makes it follow some interesting trajectory as it goes about its business. The DMP differential equations (Transformation System, Canonical System, Non-linear Function) realize a general way of generating point-to-point movements. Imitation learning using linear regression is performed to compute the weight factor W from a demonstrated trajectory dataset, given by a teacher. The quality of the imitation is evaluated by comparing the training data with the data generated by the DMP.

jponttuset.github.io

My personal webpage

gym-adv

Gym environments modified with adversarial agents

SAIL

Code for Paper "State Alignment-based Imitation Learning". Under maintenance

blender-rope-sim

fernandomayer.github.io

Source code of personal webpage

panda-env

pybullet simulated environment for panda robots, similar structure to gym

Mean-Value-Coordinates-for-Closed-Triangular-Mesh

Applications of Mean Value Coordinates for Closed Triangular Mesh