mlsanityData to 14 Oct 2026

Improving Generalization in Game Agents with Data Augmentation in Imitation Learning

arXiv 2309.12815 ↗First posted 22 Sep 2023cs.LGcited by 4

On the map: Offline RL and imitation learning from data → Imitation learning methods

Abstract (arXiv)

Imitation learning is an effective approach for training game-playing agents and, consequently, for efficient game production. However, generalization - the ability to perform well in related but unseen scenarios - is an essential requirement that remains an unsolved challenge for game AI. Generalization is difficult for imitation learning agents because it requires the algorithm to take meaningful actions outside of the training distribution. In this paper we propose a solution to this challenge. Inspired by the success of data augmentation in supervised learning, we augment the training data so the distribution of states and actions in the dataset better represents the real state-action distribution. This study evaluates methods for combining and applying data augmentations to observations, to improve generalization of imitation learning agents. It also provides a performance benchmark of these augmentations across several 3D environments. These results demonstrate that data augmentation is a promising framework for improving generalization in imitation learning agents.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Placed in Imitation learning methods, one of 13 directions in Offline RL and imitation learning from data. Builds on 24 of its 37 references that are in the map. First posted on arXiv 22 Sep 2023.

Builds on

24 of 37 references.

Closest ancestors · 5

Same direction, newest first

  1. 2022Towards Informed Design and Validation Assistance in Computer Games Using Imitation Learning15 Aug
  2. 2021Counter-Strike Deathmatch with Large-Scale Behavioural Cloning9 Apr
  3. 2020Scaling Imitation Learning in Minecraft6 Jul
  4. 2019Task-Relevant Adversarial Imitation Learning2 Oct
  5. 2018Imitation Learning with Concurrent Actions in 3D Games14 Mar

Other references · 19

  1. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,513
  2. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,183
  3. 2016Generative Adversarial Imitation Learningcited by 1,597
  4. 2019Dota 2 with Large Scale Deep Reinforcement Learningcited by 1,103
  5. 2019Unsupervised Data Augmentation for Consistency Trainingcited by 1,060
  6. 2019Leveraging Procedural Generation to Benchmark Reinforcement Learningcited by 483
  7. 2020Reinforcement Learning with Augmented Datacited by 468
  8. 2021Mastering Visual Continuous Control: Improved Data-Augmented Reinforcement Learningcited by 372
  9. 2023Imitating Human Behaviour with Diffusion Modelscited by 263
  10. 2021S4RL: Surprisingly Simple Self-Supervision for Offline Reinforcement Learningcited by 85
  11. 2018Illuminating Generalization in Deep Reinforcement Learning through Procedural Level Generationcited by 81
  12. 2021Augmenting Automated Game Testing with Deep Reinforcement Learningcited by 38
7 more
  1. 2021Improving Playtesting Coverage via Curiosity Driven Reinforcement Learning Agentscited by 28
  2. 2018Inverse reinforcement learning for video gamescited by 26
  3. 2021TeachMyAgent: a Benchmark for Automatic Curriculum Learning in Deep RLcited by 18
  4. 2019Winning Isn't Everything: Enhancing Game Development with Intelligent Agentscited by 15
  5. 2020DeepCrawl: Deep Reinforcement Learning for Turn-based Strategy Gamescited by 6
  6. 2020Demonstration-efficient Inverse Reinforcement Learning in Procedurally Generated Environmentscited by 6
  7. 2022CCPT: Automatic Gameplay Testing and Validation with Curiosity-Conditioned Proximal Trajectoriescited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 1; 2025: 1; 2026: 2.

4 papers.

  1. 2024Efficient Active Imitation Learning with Random Network Distillationcited by 7
  2. 2026Towards trustworthy agentic AI: a comprehensive survey of safety, robustness, privacy, and system securitycited by 5
  3. 2025Learning Representations in Video Game Agents with Supervised Contrastive Imitation Learningcited by 1
  4. 2026Augmentations for Robust and Efficient Imitation Learning in Streamed Video Gamescited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023Guided Data Augmentation for Offline Reinforcement Learning and Imitation Learning
  2. 2021Adversarial Imitation Learning with Trajectorial Augmentation and Correction
  3. 2025Generalization Capability for Imitation Learning
  4. 2024Imitation Learning Datasets: A Toolkit For Creating Datasets, Training Agents and Benchmarking
  5. 2025Imitation Learning in the Deep Learning Era: A Novel Taxonomy and Recent Advances
  6. 2025Quantifying Generalisation in Imitation Learning
  7. 2024A Recipe for Unbounded Data Augmentation in Visual Reinforcement Learning
  8. 2020Generalization Guarantees for Imitation Learning