This repo includes several tasks/environments built on openai-gym for training RL agents.

The original paper: Martinolli, Marco, Wulfram Gerstner, and Aditya Gilra. ‘Multi-Timescale Memory Dynamics Extend Task Repertoire in a Reinforcement Learning Network With Attention-Gated Memory’. Frontiers in Computational Neuroscience 12 (2018).

Install

$ git clone https://github.com/CgnRLAgent/cog_ml_tasks.git
$ cd cog_ml_tasks
$ pip install -e .

Requirements: see the setup.py.

Envs/Tasks Descriptions

1. 12_AX
2. AX_CPT
3. 12_AX_S
4. 12_AX_CPT
5. Simple_Copy
6. Simple_Copy_Repeat
7. Sequence Prediction
8. Saccade