hcRL

Code and data for the paper:
"Hippocampal networks support reinforcement learning in partially observable environments"

---

1. System Requirements

• Operating System: Tested on Ubuntu 20.04
• Python version: Python 3.9
• Hardware: GPU (CUDA-compatible), GeForce RTX 2080 Ti.

Software Dependencies

Package	Version
Python	3.9
PyTorch	≥ 1.12.0
Hydra	≥ 1.2.0
wandb	≥ 0.18.7
numpy	≥ 1.24
gym	≥ 0.26.2
omegaconf	≥ 2.2.2
minigrid	≥ 2.3.1

---

2. Installation Guide

Step 1: Clone the repository

git clone https://github.com/neuralml/hcRL.git
cd hcRL

Step 2: Create and activate the Conda environment

conda env create -n hcRL
conda activate hcRL

3. Instructions

To run a full training session with your own configuration:

python train.py agent=dqn seed=1 experiment=exp use_wandb=true

To train hcDRQN run

python train.py agent=drqn seed=1 experiment=exp use_wandb=true

Results are logged and tracked via Weights & Biases, if enabled.

4. License

This project is licensed under the MIT License. See LICENSE for details.