mlsanityData to 7 Oct 2026

Investigating Navigation Strategies in the Morris Water Maze through Deep Reinforcement Learning

arXiv 2306.01066 ↗First posted 1 Jun 2023cs.LG cs.NE q-bio.NCcited by 3

Not on the map. Why?

Abstract (arXiv)

Navigation is a complex skill with a long history of research in animals and humans. In this work, we simulate the Morris Water Maze in 2D to train deep reinforcement learning agents. We perform automatic classification of navigation strategies, analyze the distribution of strategies used by artificial agents, and compare them with experimental data to show similar learning dynamics as those seen in humans and rodents. We develop environment-specific auxiliary tasks and examine factors affecting their usefulness. We suggest that the most beneficial tasks are potentially more biologically feasible for real agents to use. Lastly, we explore the development of internal representations in the activations of artificial agent neural networks. These representations resemble place cells and head-direction cells found in mouse brains, and their presence has correlation to the navigation strategies that artificial agents employ.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 22 of its 43 references that are in the map. First posted on arXiv 1 Jun 2023.

Builds on

22 of 43 references.

References · 22

  1. 2017Proximal Policy Optimization Algorithmscited by 14,441
  2. 2022A ConvNet for the 2020scited by 3,545
  3. 2016Asynchronous Methods for Deep Reinforcement Learningcited by 2,218
  4. 2016On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minimacited by 1,337
  5. 2014On the Properties of Neural Machine Translation: Encoder-Decoder Approachescited by 1,303
  6. 2017Curiosity-driven Exploration by Self-supervised Predictioncited by 1,173
  7. 2019Dota 2 with Large Scale Deep Reinforcement Learningcited by 1,084
  8. 2016Reinforcement Learning with Unsupervised Auxiliary Taskscited by 400
  9. 2016Target-driven Visual Navigation in Indoor Scenes using Deep Reinforcement Learningcited by 351
  10. 2018An Empirical Model of Large-Batch Trainingcited by 268
  11. 2016Learning to Navigate in Complex Environmentscited by 192
  12. 2020fastai: A Layered API for Deep Learningcited by 183
10 more
  1. 2016ViZDoom: A Doom-based AI Research Platform for Visual Reinforcement Learningcited by 178
  2. 2018Adapting Auxiliary Losses Using Gradient Similaritycited by 98
  3. 2016Playing FPS Games with Deep Reinforcement Learningcited by 95
  4. 2016Deep Successor Reinforcement Learningcited by 72
  5. 2021On The Effect of Auxiliary Tasks on Representation Dynamicscited by 56
  6. 2019A Geometric Perspective on Optimal Representations for Reinforcement Learningcited by 54
  7. 2018Accelerated Methods for Deep Reinforcement Learningcited by 40
  8. 2021Large Batch Simulation for Deep Reinforcement Learningcited by 18
  9. 2019Terminal Prediction as an Auxiliary Task for Deep Reinforcement Learningcited by 11
  10. 2017A generalised framework for detailed classification of swimming paths inside the Morris Water Mazecited by 2

Built on by

Cited by year

Citing papers per year: 2024: 3.

3 papers.

  1. 2024Transformers Can Navigate Mazes With Multi-Step Predictioncited by 11
  2. 2024A Role of Environmental Complexity on Representation Learning in Deep Reinforcement Learning Agentscited by 2
  3. 2024A transformer-based deep reinforcement learning approach to spatial navigation in a partially observable Morris Water Mazecited by 0