# Fostering Intrinsic Motivation in Reinforcement Learning with Pretrained Foundation Models

arXiv: https://arxiv.org/abs/2410.07404 · HTML: https://mlsanity.com/paper/2410.07404/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Deep RL algorithms and representations, one of 15 directions in Deep reinforcement learning methods.
- Builds on 23 of its 27 references that are in the map.
- First posted on arXiv 9 Oct 2024.

## Placed in

- Deep reinforcement learning methods / [Deep RL algorithms and representations](https://mlsanity.com/deeprl/deep-rl-algorithms-and-representations/)

## Builds on (most cited first, 23 of 23)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [Curiosity-driven Exploration by Self-supervised Prediction](https://mlsanity.com/paper/1705.05363/) (2017)
- [Exploration by Random Network Distillation](https://mlsanity.com/paper/1810.12894/) (2018)
- [Unifying Count-Based Exploration and Intrinsic Motivation](https://mlsanity.com/paper/1606.01868/) (2016)
- [Noisy Networks for Exploration](https://mlsanity.com/paper/1706.10295/) (2017)
- [Count-Based Exploration with Neural Density Models](https://mlsanity.com/paper/1703.01310/) (2017)
- [Guiding Pretraining in Reinforcement Learning with Large Language Models](https://mlsanity.com/paper/2302.06692/) (2023)
- [Vision-Language Models are Zero-Shot Reward Models for Reinforcement Learning](https://mlsanity.com/paper/2310.12921/) (2023)
- [RIDE: Rewarding Impact-Driven Exploration for Procedurally-Generated Environments](https://mlsanity.com/paper/2002.12292/) (2020)
- [Episodic Curiosity through Reachability](https://mlsanity.com/paper/1810.02274/) (2018)
- [Learning with AMIGo: Adversarially Motivated Intrinsic Goals](https://mlsanity.com/paper/2006.12122/) (2020)
- [Vision-Language Models as a Source of Rewards](https://mlsanity.com/paper/2312.09187/) (2023)
- [Imitation Learning: Progress, Taxonomies and Challenges](https://mlsanity.com/paper/2106.12177/) (2021)
- [Adversarially Guided Actor-Critic](https://mlsanity.com/paper/2102.04376/) (2021)
- [Language Reward Modulation for Pretraining Reinforcement Learning](https://mlsanity.com/paper/2308.12270/) (2023)
- [FuRL: Visual-Language Models as Fuzzy Rewards for Reinforcement Learning](https://mlsanity.com/paper/2406.00645/) (2024)
- [Accelerating Reinforcement Learning of Robotic Manipulations via Feedback from Large Language Models](https://mlsanity.com/paper/2311.02379/) (2023)
- [A Study of Global and Episodic Bonuses for Exploration in Contextual MDPs](https://mlsanity.com/paper/2306.03236/) (2023)
- [LECO: Learnable Episodic Count for Task-Specific Intrinsic Reward](https://mlsanity.com/paper/2210.05409/) (2022)
- [Foundation Models for Semantic Novelty in Reinforcement Learning](https://mlsanity.com/paper/2211.04878/) (2022)
- [An Evaluation Study of Intrinsic Motivation Techniques applied to Reinforcement Learning over Hard Exploration Environments](https://mlsanity.com/paper/2205.11184/) (2022)

## Abstract (arXiv)

Exploration remains a significant challenge in reinforcement learning, especially in environments where extrinsic rewards are sparse or non-existent. The recent rise of foundation models, such as CLIP, offers an opportunity to leverage pretrained, semantically rich embeddings that encapsulate broad and reusable knowledge. In this work we explore the potential of these foundation models not just to drive exploration, but also to analyze the critical role of the episodic novelty term in enhancing exploration effectiveness of the agent. We also investigate whether providing the intrinsic module with complete state information -- rather than just partial observations -- can improve exploration, despite the difficulties in handling small variations within large state spaces. Our experiments in the MiniGrid domain reveal that intrinsic modules can effectively utilize full state information, significantly increasing sample efficiency while learning an optimal policy. Moreover, we show that the embeddings provided by foundation models are sometimes even better than those constructed by the agent during training, further accelerating the learning process, especially when coupled with the episodic novelty term to enhance exploration.