mlsanityData to 14 Oct 2026

Reasoning-Modulated Representations

arXiv 2107.08881 ↗First posted 19 Jul 2021cs.LG cs.AI stat.MLcited by 13

Not on the map. Why?

Abstract (arXiv)

Neural networks leverage robust internal representations in order to generalise. Learning them is difficult, and often requires a large training set that covers the data distribution densely. We study a common setting where our task is not purely opaque. Indeed, very often we may have access to information about the underlying system (e.g. that observations must obey certain laws of physics) that any "tabula rasa" neural network would need to re-learn from scratch, penalising performance. We incorporate this information into a pre-trained reasoning module, and investigate its role in shaping the discovered representations in diverse self-supervised learning settings from pixels. Our approach paves the way for a new class of representation learning, grounded in algorithmic priors.

Key facts. Cited by 13 papers in the map (data through 14 Oct 2026). Builds on 46 of its 61 references that are in the map. First posted on arXiv 19 Jul 2021.

Builds on

46 of 61 references.

References · 46

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2020Language Models are Few-Shot Learnerscited by 32,419
  3. 2017Neural Message Passing for Quantum Chemistrycited by 3,487
  4. 2018World Modelscited by 1,513
  5. 2017Deep Setscited by 1,421
  6. 2019Mastering Atari, Go, Chess and Shogi by Planning with a Learned Modelcited by 1,353
  7. 2018Reconciling modern machine learning practice and the bias-variance trade-offcited by 927
  8. 2017Revisiting Unreasonable Effectiveness of Data in Deep Learning Eracited by 919
  9. 2020Learning to Simulate Complex Physics with Graph Networkscited by 880
  10. 2020Object-Centric Learning with Slot Attentioncited by 876
  11. 2020Learning Mesh-Based Simulation with Graph Networkscited by 868
  12. 2016Interaction Networks for Learning about Objects, Relations and Physicscited by 509
34 more
  1. 2017A simple neural network module for relational reasoningcited by 346
  2. 2020Discovering Symbolic Models from Deep Learning with Inductive Biasescited by 344
  3. 2018Neural Relational Inference for Interacting Systemscited by 339
  4. 2020Agent57: Outperforming the Atari Human Benchmarkcited by 303
  5. 2021Combinatorial optimization and reasoning with graph neural networkscited by 268
  6. 2020On the Binding Problem in Artificial Neural Networkscited by 233
  7. 2020How Neural Networks Extrapolate: From Feedforward to Graph Neural Networkscited by 222
  8. 2019Recurrent Independent Mechanismscited by 219
  9. 2019Contrastive Learning of Structured World Modelscited by 209
  10. 2019What Can Neural Networks Reason About?cited by 159
  11. 2019Unsupervised State Representation Learning in Ataricited by 144
  12. 2018Sequential Attend, Infer, Repeat: Generative Modelling of Moving Objectscited by 128
  13. 2019Neural Execution of Graph Algorithmscited by 121
  14. 2016A Compositional Object-Based Approach to Learning Physical Dynamicscited by 119
  15. 2019Spatial Broadcast Decoder: A Simple Architecture for Learning Disentangled Representations in VAEscited by 117
  16. 2019Entity Abstraction in Visual Model-Based Reinforcement Learningcited by 110
  17. 2019Unsupervised Learning of Object Keypoints for Perception and Controlcited by 109
  18. 2019Hamiltonian Graph Networks with ODE Integratorscited by 108
  19. 2019SCALOR: Generative World Models with Scalable Object Representationscited by 108
  20. 2018Relational Neural Expectation Maximization: Unsupervised Discovery of Objects and their Interactionscited by 100
  21. 2017Visual Interaction Networkscited by 95
  22. 2021Neural Algorithmic Reasoningcited by 94
  23. 2020Causal Discovery in Physical Systems from Videoscited by 77
  24. 2021Neural Production Systems: Learning Rule-Governed Visual Dynamicscited by 75
  25. 2020Improving Generative Imagination in Object-Centric World Modelscited by 71
  26. 2018Relational recurrent neural networkscited by 65
  27. 2020Pointer Graph Networkscited by 53
  28. 2019Physics-as-Inverse-Graphics: Unsupervised Physical Parameter Estimation from Videocited by 45
  29. 2018Relational inductive bias for physical construction in humans and machinescited by 44
  30. 2018Combined Reinforcement Learning via Abstract Representationscited by 37
  31. 2020Learning Object-Centric Video Models by Contrasting Setscited by 19
  32. 2020XLVIN: eXecuted Latent Value Iteration Netscited by 17
  33. 2021Persistent Message Passingcited by 13
  34. 2020Towards Scale-Invariant Graph-related Problem Solving by Iterative Homogeneous Graph Neural Networkscited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 4; 2023: 3; 2024: 4; 2025: 1; 2026: 1.

13 papers.

Sort:

By area:

  1. 2022The CLRS Algorithmic Reasoning Benchmarkcited by 106
  2. 2022A Generalist Neural Algorithmic Learnercited by 59
  3. 2022Graph Neural Networks are Dynamic Programmerscited by 55
  4. 2022Learning Symmetric Embeddings for Equivariant World Modelscited by 54
  5. 2023Neural Algorithmic Reasoning with Causal Regularisationcited by 31
  6. 2023Dual Algorithmic Reasoningcited by 21
  7. 2024Learning Physics From Video: Unsupervised Physical Parameter Estimation for Continuous Dynamical Systemscited by 13
  8. 2024On the Markov Property of Neural Algorithmic Reasoning: Analyses and Methodscited by 11
  9. 2023Parallel Algorithms Align with Neural Executioncited by 6
  10. 2024Recurrent Aggregators in Neural Algorithmic Reasoningcited by 6
  11. 2026IRIS: A Real-World Benchmark for Inverse Recovery and Identification of Physical Dynamic Systems from Monocular Videocited by 4
  12. 2024Open-Book Neural Algorithmic Reasoningcited by 3
1 more
  1. 2025Leveraging Classical Algorithms for Graph Neural Networkscited by 0