mlsanityData to 7 Oct 2026

Bridging Imagination and Reality for Model-Based Deep Reinforcement Learning

arXiv 2010.12142 ↗First posted 23 Oct 2020cs.LGcited by 15

On the map: World models → Latent dynamics models and model-based RL

Abstract (arXiv)

Sample efficiency has been one of the major challenges for deep reinforcement learning. Recently, model-based reinforcement learning has been proposed to address this challenge by performing planning on imaginary trajectories with a learned world model. However, world model learning may suffer from overfitting to training trajectories, and thus model-based value estimation and policy search will be pone to be sucked in an inferior local policy. In this paper, we propose a novel model-based reinforcement learning algorithm, called BrIdging Reality and Dream (BIRD). It maximizes the mutual information between imaginary and real trajectories so that the policy improvement learned from imaginary trajectories can be easily generalized to real trajectories. We demonstrate that our approach improves sample efficiency of model-based planning, and achieves state-of-the-art performance on challenging visual control benchmarks.

Key facts. Cited by 15 papers in the map (data through 7 Oct 2026). Placed in Latent dynamics models and model-based RL, one of 11 directions in World models. Builds on 9 of its 65 references that are in the map. First posted on arXiv 23 Oct 2020.

Builds on

9 of 65 references.

References · 9

  1. 2017Proximal Policy Optimization Algorithmscited by 14,441
  2. 2013Auto-Encoding Variational Bayescited by 3,980
  3. 2019Mastering Atari, Go, Chess and Shogi by Planning with a Learned Modelcited by 1,327
  4. 2019Model-Based Reinforcement Learning for Ataricited by 495
  5. 2018Model-Based Value Estimation for Efficient Model-Free Reinforcement Learningcited by 152
  6. 2020Model-based Reinforcement Learning: A Surveycited by 39
  7. 2018Universal Planning Networkscited by 29
  8. 2019Imagined Value Gradients: Model-Based Policy Optimization with Transferable Latent Dynamics Modelscited by 26
  9. 2017Model-Based Planning with Discrete and Continuous Actionscited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 5; 2022: 4; 2023: 0; 2024: 2; 2025: 2; 2026: 2.

15 papers.

Sort:

By area:

  1. 2022Enhance Sample Efficiency and Robustness of End-to-end Urban Autonomous Driving via Semantic Masked World Modelcited by 44
  2. 2022Flow-based Recurrent Belief State Learning for POMDPscited by 16
  3. 2021On Effective Scheduling of Model-based Reinforcement Learningcited by 13
  4. 2021Improving Generalization in Meta-RL with Imaginary Tasks from Latent Dynamics Mixturecited by 12
  5. 2025DMWM: Dual-Mind World Model with Long-Term Imaginationcited by 10
  6. 2022The developmental trajectory of object recognition robustness: children are like small adults but unlike big deep neural networkscited by 9
  7. 2026Biased Dreams: Limitations to Epistemic Uncertainty Quantification in Latent Dynamics Modelscited by 5
  8. 2021Acting upon Imagination: when to trust imagined trajectories in model based reinforcement learningcited by 3
  9. 2021Uncertainty-Aware Model-Based Reinforcement Learning with Application to Autonomous Drivingcited by 3
  10. 2024Model-Driven Deep Neural Network for Enhanced AoA Estimation Using 5G gNBcited by 2
  11. 2025Ego-centric Learning of Communicative World Models for Autonomous Drivingcited by 2
  12. 2021ED2: Environment Dynamics Decomposition World Models for Continuous Controlcited by 0
3 more
  1. 2022Model-based Reinforcement Learning with a Hamiltonian Canonical ODE Networkcited by 0
  2. 2024Do Agents Dream of Electric Sheep?: Improving Generalization in Reinforcement Learning through Generative Learningcited by 0
  3. 2026Improving Generalization and Robustness in Offline Reinforcement Learning via Boundary-Aware Data Augmentationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2019Dream to Control: Learning Behaviors by Latent Imagination
  2. 2026Dream-MPC: Gradient-Based Model Predictive Control with Latent Imagination
  3. 2020Dreaming: Model-based Reinforcement Learning by Latent Imagination without Reconstruction
  4. 2023Model-Based Reinforcement Learning with Isolated Imaginations
  5. 2026Q-Learning With World Models
  6. 2021Latent Imagination Facilitates Zero-Shot Transfer in Autonomous Racing
  7. 2026GIRL: Generative Imagination Reinforcement Learning via Information-Theoretic Hallucination Control
  8. 2018Learning Real-World Robot Policies by Dreaming