iQRL: implicitly Quantized Representations for Sample-efficient Reinforcement Learning

This repository is the official implementation of iQRL, a reinforcement learning algorithm for continuous control.

Learning representations for reinforcement learning (RL) has shown much promise for continuous control. We propose an efficient representation learning method using only a self-supervised latent-state consistency loss. Our approach employs an encoder and a dynamics model to map observations to latent states and predict future latent states, respectively. We achieve high performance and prevent representation collapse by quantizing the latent representation such that the rank of the representation is empirically preserved. Our method, named iQRL: implicitly Quantized Reinforcement Learning, is straightforward, compatible with any model-free RL algorithm, and demonstrates excellent performance by outperforming other recently proposed representation learning methods in continuous control benchmarks from DeepMind Control Suite.

Install instructions

Install dependencies:

conda env create -f environment.yml
conda activate iqrl

You might need to install PyTorch with CUDA/ROCm.

Running experiments

Train the agent:

python train.py +env=walker-walk

To log metrics with W&B:

python train.py +env=walker-walk ++use_wandb=True

All tested tasks are listed incfgs/env.

Configuring experiments

This repo uses hydra for configuration. You can easily try new hyperparameters for iQRL by overriding them on the command line. For example,

python train.py +env=walker-walk ++use_wandb=True ++agent.batch_size=512

changes the batch size to be 512 instead of the default value found in iqrl.py/iQRLConfig.

You can also use hydra to submit multiple Slurm jobs directly from the command line using

python train.py -m +env=walker-walk ++use_wandb=True ++agent.batch_size=256,512 ++agent.lr=1e-4,1e-4

This uses the utils/cluster_utils.py/SlurmConfig to configure the jobs, setting timeout_min=1440 (i.e. 24hrs) and mem_gb=32. If you want to run the job for longer (e.g 48hrs), you can use the following

python train.py -m +env=walker-walk ++use_wandb=True ++agent.batch_size=256,512 ++agent.lr=1e-4,1e-4 ++hydra.launcher.timeout_min=2880

BibTeX

Please consider citing our arXiv paper:

@misc{scannell2024iqrl,
  title           = {iQRL - Implicitly Quantized Representations for Sample-efficient Reinforcement Learning},
  author          = {Aidan Scannell and Kalle Kujanpää and Yi Zhao and Mohammadreza Nakhaei and Arno Solin and Joni Pajarinen},
  year            = {2024},
  eprint          = {2406.02696},
  archivePrefix   = {arXiv},
  primaryClass    = {cs.LG}
}

and our workshop paper:

@inproceedings{scannellQuantized2024,
  title           = {Quantized Representations Prevent Dimensional Collapse in Self-predictive {RL}},
  booktitle       = {ICML Workshop on Aligning Reinforcement Learning Experimentalists and Theorists ({ARLET})},
  author          = {Aidan Scannell and Kalle Kujanpää and Yi Zhao and Mohammadreza Nakhaei and Arno Solin and Joni Pajarinen},
  year            = {2024},
}