mlsanityData to 7 Oct 2026

Snowflake: Scaling GNNs to High-Dimensional Continuous Control via Parameter Freezing

arXiv 2103.01009 ↗First posted 1 Mar 2021cs.LGcited by 7

Not on the map. Why?

Abstract (arXiv)

Recent research has shown that graph neural networks (GNNs) can learn policies for locomotion control that are as effective as a typical multi-layer perceptron (MLP), with superior transfer and multi-task performance (Wang et al., 2018; Huang et al., 2020). Results have so far been limited to training on small agents, with the performance of GNNs deteriorating rapidly as the number of sensors and actuators grows. A key motivation for the use of GNNs in the supervised learning setting is their applicability to large graphs, but this benefit has not yet been realised for locomotion control. We identify the weakness with a common GNN architecture that causes this poor scaling: overfitting in the MLPs within the network that encode, decode, and propagate messages. To combat this, we introduce Snowflake, a GNN training method for high-dimensional continuous control that freezes parameters in parts of the network that suffer from overfitting. Snowflake significantly boosts the performance of GNNs for locomotion control on large agents, now matching the performance of MLPs, and with superior transfer properties.

Key facts. Cited by 7 papers in the map (data through 7 Oct 2026). Builds on 41 of its 65 references that are in the map. First posted on arXiv 1 Mar 2021.

Builds on

41 of 65 references.

References · 41

  1. 2017Attention Is All You Needcited by 59,932
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2017Proximal Policy Optimization Algorithmscited by 14,441
  4. 2013Efficient Estimation of Word Representations in Vector Spacecited by 4,346
  5. 2019Parameter-Efficient Transfer Learning for NLPcited by 4,195
  6. 2014Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translationcited by 4,016
  7. 2017Neural Message Passing for Quantum Chemistrycited by 3,460
  8. 2015Trust Region Policy Optimizationcited by 2,503
  9. 2015High-Dimensional Continuous Control Using Generalized Advantage Estimationcited by 1,916
  10. 2019A Survey on Bias and Fairness in Machine Learningcited by 1,911
  11. 2019SuperGlue: Learning Feature Matching with Graph Neural Networkscited by 1,437
  12. 2018Relational inductive biases, deep learning, and graph networkscited by 1,341
29 more
  1. 2014How transferable are features in deep neural networks?cited by 1,216
  2. 2015Gated Graph Sequence Neural Networkscited by 792
  3. 2013Exact solutions to the nonlinear dynamics of learning in deep linear neural networkscited by 748
  4. 2017Learning Combinatorial Optimization Algorithms over Graphscited by 557
  5. 2018Maximum a Posteriori Policy Optimisationcited by 318
  6. 2018Graph networks as learnable physics engines for inference and controlcited by 242
  7. 2013Bayesian Optimization in a Billion Dimensions via Random Embeddingscited by 178
  8. 2017Emergence of Locomotion Behaviours in Rich Environmentscited by 159
  9. 2016Value Iteration Networkscited by 158
  10. 2016Discriminative Embeddings of Latent Variable Models for Structured Datacited by 130
  11. 2020One Policy to Control Them All: Shared Modular Policies for Agent-Agnostic Controlcited by 114
  12. 2016Fairness in Reinforcement Learningcited by 93
  13. 2019Learning to Control Self-Assembling Morphologies: A Study of Generalization via Modularitycited by 68
  14. 2020My Body is a Cage: the Role of Morphology in Graph-Based Incompatible Controlcited by 68
  15. 2020Deep Implicit Coordination Graphs for Multi-agent Reinforcement Learningcited by 58
  16. 2017FreezeOut: Accelerate Training by Progressively Freezing Layerscited by 54
  17. 2019Structured agents for physical constructioncited by 50
  18. 2018Person Re-identification with Deep Similarity-Guided Graph Neural Networkcited by 49
  19. 2020Fixed Encoder Self-Attention Patterns in Transformer-Based Machine Translationcited by 42
  20. 2020Hard-Coded Gaussian Attention for Neural Machine Translationcited by 42
  21. 2020Randomized Entity-wise Factorization for Multi-Agent Reinforcement Learningcited by 38
  22. 2019Working Memory Graphscited by 23
  23. 2020Accelerated Charged Particle Tracking with Graph Neural Networks on FPGAscited by 23
  24. 2017Policy Gradient Methods for Reinforcement Learning with Function Approximation and Action-Dependent Baselinescited by 20
  25. 2020XLVIN: eXecuted Latent Value Iteration Netscited by 17
  26. 2020Contextual Embeddings: When Are They Worth It?cited by 15
  27. 2020Reward Propagation Using Graph Convolutional Networkscited by 15
  28. 2017An Exploration of Word Embedding Initialization in Deep-Learning Taskscited by 6
  29. 2017A Comparative Study of Word Embeddings for Reading Comprehensioncited by 4

Built on by

Cited by year

Citing papers per year: 2022: 3; 2023: 2; 2024: 2.

7 papers.

  1. 2022MetaMorph: Learning Universal Controllers with Transformerscited by 78
  2. 2022A Generalist Neural Algorithmic Learnercited by 58
  3. 2023LightTS: Lightweight Time Series Classification with Adaptive Ensemble Distillation -- Extended Versioncited by 49
  4. 2023A Generalist Dynamics Model for Controlcited by 24
  5. 2022A System for Morphology-Task Generalization via Unified Representation and Behavior Distillationcited by 12
  6. 2024Subequivariant Reinforcement Learning Framework for Coordinated Motion Controlcited by 2
  7. 2024MeMo: Meaningful, Modular Controllers via Noise Injectioncited by 1