mlsanityData to 14 Oct 2026

The Free Energy Principle for Perception and Action: A Deep Learning Perspective

arXiv 2207.06415 ↗First posted 13 Jul 2022cs.LG cs.AI q-bio.NCcited by 24

Not on the map. Why?

Abstract (arXiv)

The free energy principle, and its corollary active inference, constitute a bio-inspired theory that assumes biological agents act to remain in a restricted set of preferred states of the world, i.e., they minimize their free energy. Under this principle, biological agents learn a generative model of the world and plan actions in the future that will maintain the agent in an homeostatic state that satisfies its preferences. This framework lends itself to being realized in silico, as it comprehends important aspects that make it computationally affordable, such as variational inference and amortized planning. In this work, we investigate the tool of deep learning to design and realize artificial agents based on active inference, presenting a deep-learning oriented presentation of the free energy principle, surveying works that are relevant in both machine learning and active inference areas, and discussing the design choices that are involved in the implementation process. This manuscript probes newer perspectives for the active inference framework, grounding its theoretical aspects into more pragmatic affairs, offering a practical guide to active inference newcomers and a starting point for deep learning practitioners that would like to investigate implementations of the free energy principle.

Key facts. Cited by 24 papers in the map (data through 14 Oct 2026). Builds on 89 of its 165 references that are in the map. First posted on arXiv 13 Jul 2022.

Builds on

89 of 165 references.

References · 89

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2017Proximal Policy Optimization Algorithmscited by 15,215
  4. 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
  5. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,885
  6. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,750
  7. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
  8. 2013Auto-Encoding Variational Bayescited by 4,053
  9. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,926
  10. 2013Playing Atari with Deep Reinforcement Learningcited by 3,274
  11. 2020Exploring Simple Siamese Representation Learningcited by 2,866
  12. 2020Unsupervised Learning of Visual Features by Contrasting Cluster Assignmentscited by 2,857
38 more
  1. 2014Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modelingcited by 2,367
  2. 2015Variational Inference with Normalizing Flowscited by 2,170
  3. 2013Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computationcited by 2,079
  4. 2021Barlow Twins: Self-Supervised Learning via Redundancy Reductioncited by 1,838
  5. 2019Dream to Control: Learning Behaviors by Latent Imaginationcited by 1,653
  6. 2016Variational Inference: A Review for Statisticianscited by 1,472
  7. 2015Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcastingcited by 1,436
  8. 2019Generating Diverse High-Fidelity Images with VQ-VAE-2cited by 1,387
  9. 2019Mastering Atari, Go, Chess and Shogi by Planning with a Learned Modelcited by 1,353
  10. 2018Learning Latent Dynamics for Planning from Pixelscited by 1,264
  11. 2017Curiosity-driven Exploration by Self-supervised Predictioncited by 1,194
  12. 2021Alias-Free Generative Adversarial Networkscited by 1,184
  13. 2017Hindsight Experience Replaycited by 1,136
  14. 2020Mastering Atari with Discrete World Modelscited by 1,007
  15. 2016Deep Variational Information Bottleneckcited by 930
  16. 2018Recurrent World Models Facilitate Policy Evolutioncited by 785
  17. 2020CURL: Contrastive Unsupervised Representations for Reinforcement Learningcited by 760
  18. 2017Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithmcited by 734
  19. 2019Solving Rubik's Cube with a Robot Handcited by 734
  20. 2020NVAE: A Deep Hierarchical Variational Autoencodercited by 637
  21. 2015Weight Uncertainty in Neural Networkscited by 554
  22. 2019Model-Based Reinforcement Learning for Ataricited by 502
  23. 2018Reinforcement Learning and Control as Probabilistic Inference: Tutorial and Reviewcited by 482
  24. 2021Brax -- A Differentiable Physics Engine for Large Scale Rigid Body Simulationcited by 407
  25. 2016The CMA Evolution Strategy: A Tutorialcited by 377
  26. 2020Planning to Explore via Self-Supervised World Modelscited by 368
  27. 2020Data-Efficient Reinforcement Learning with Self-Predictive Representationscited by 318
  28. 2020Agent57: Outperforming the Atari Human Benchmarkcited by 303
  29. 2018Large-Scale Study of Curiosity-Driven Learningcited by 280
  30. 2019Self-Supervised Exploration via Disagreementcited by 265
  31. 2018Stochastic Video Generation with a Learned Priorcited by 245
  32. 2019Stochastic Latent Actor-Critic: Deep Reinforcement Learning with a Latent Variable Modelcited by 231
  33. 2020Concept Whitening for Interpretable Image Recognitioncited by 225
  34. 2016Deep Predictive Coding Networks for Video Prediction and Unsupervised Learningcited by 204
  35. 2018Sensitivity and Generalization in Neural Networks: an Empirical Studycited by 189
  36. 2019Efficient Exploration via State Marginal Matchingcited by 188
  37. 2021Maximum Entropy RL (Provably) Solves Some Robust RL Problemscited by 176
  38. 2019Hamiltonian Generative Networkscited by 137

50 most cited of 89 All 89 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 2; 2023: 8; 2024: 5; 2025: 5; 2026: 4.

24 papers.

Sort:

By area:

  1. 2022Designing Ecosystems of Intelligence from First Principlescited by 23
  2. 2023On efficient computation in active inferencecited by 15
  3. 2024R-AIF: Solving Sparse-Reward Robotic Tasks from Pixels with Active Inference and World Modelscited by 10
  4. 2023Supervised structure learningcited by 9
  5. 2022Disentangling Shape and Pose for Object-Centric Deep Active Inference Modelscited by 6
  6. 2023Resolving uncertainty on the fly: Modeling adaptive driving behavior as active inferencecited by 6
  7. 2023Symmetry and Complexity in Object-Centric Deep Active Inference Modelscited by 4
  8. 2023Latent Noise Segmentation: How Neural Noise Leads to the Emergence of Segmentation and Groupingcited by 4
  9. 2024Value of Information and Reward Specification in Active Inference and POMDPscited by 4
  10. 2024Possible Principles for Aligned Structure Learning Agentscited by 3
  11. 2025Active inference and artificial reasoningcited by 3
  12. 2023Rejecting Cognitivism: Computational Phenomenology for Deep Learningcited by 2
12 more
  1. 2024Free Energy Projective Simulation (FEPS): Active inference with interpretabilitycited by 2
  2. 2025EASE: Embodied Active Event Perception via Self-Supervised Energy Minimizationcited by 2
  3. 2026The SIGReg Objective as Variational Free Energy: A Theoretical Active-Inference Account of JEPA World Modelscited by 2
  4. 2023Advancing Perception in Artificial Intelligence through Principles of Cognitive Sciencecited by 1
  5. 2024Toward Universal and Interpretable World Models for Open-ended Learning Agentscited by 1
  6. 2025Prior preferences in active inference agents: soft, hard, and goal shapingcited by 1
  7. 2026Active Inference as a Convex Markov Decision Processcited by 1
  8. 2023A Neural Network Implementation for Free Energy Principlecited by 0
  9. 2025The Physical Basis of Prediction: World Model Formation in Neural Organoids via an LLM-Generated Curriculumcited by 0
  10. 2025Real-World Robot Control by Deep Active Inference With a Temporally Hierarchical World Modelcited by 0
  11. 2026Optimizing Neurorobot Policy under Limited Demonstration Data through Preference Regretcited by 0
  12. 2026On Distinguishing Capability Elicitation from Capability Creation in Post-Training: A Free-Energy Perspectivecited by 0