mlsanityData to 7 Oct 2026

Multipath agents for modular multitask ML systems

arXiv 2302.02721 ↗First posted 6 Feb 2023cs.LG cs.AIcited by 1

Not on the map. Why?

Abstract (arXiv)

A standard ML model is commonly generated by a single method that specifies aspects such as architecture, initialization, training data and hyperparameters configuration. The presented work introduces a novel methodology allowing to define multiple methods as distinct agents. Agents can collaborate and compete to generate and improve ML models for a given tasks. The proposed methodology is demonstrated with the generation and extension of a dynamic modular multitask ML system solving more than one hundred image classification tasks. Diverse agents can compete to produce the best performing model for a task by reusing the modules introduced to the system by competing agents. The presented work focuses on the study of agents capable of: 1) reusing the modules generated by concurrent agents, 2) activating in parallel multiple modules in a frozen state by connecting them with trainable modules, 3) condition the activation mixture on each data sample by using a trainable router module. We demonstrate that this simple per-sample parallel routing method can boost the quality of the combined solutions by training a fraction of the activated parameters.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 51 of its 62 references that are in the map. First posted on arXiv 6 Feb 2023.

Builds on

51 of 62 references.

References · 51

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  3. 2020Language Models are Few-Shot Learnerscited by 32,199
  4. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  5. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  6. 2020Scaling Laws for Neural Language Modelscited by 6,411
  7. 2022Flamingo: a Visual Language Model for Few-Shot Learningcited by 4,337
  8. 2019Parameter-Efficient Transfer Learning for NLPcited by 4,195
  9. 2017Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layercited by 3,413
  10. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,708
  11. 2016Learning without Forgettingcited by 2,465
  12. 2018DARTS: Differentiable Architecture Searchcited by 1,782
38 more
  1. 2016Neural Architecture Search with Reinforcement Learningcited by 1,489
  2. 2017Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semanticscited by 1,345
  3. 2016Progressive Neural Networkscited by 1,332
  4. 2022LaMDA: Language Models for Dialog Applicationscited by 1,218
  5. 2020Gradient Surgery for Multi-Task Learningcited by 1,210
  6. 2018MnasNet: Platform-Aware Neural Architecture Search for Mobilecited by 937
  7. 2018Regularized Evolution for Image Classifier Architecture Searchcited by 902
  8. 2021GLaM: Efficient Scaling of Language Models with Mixture-of-Expertscited by 890
  9. 2018Efficient Neural Architecture Search via Parameter Sharingcited by 831
  10. 2017GradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networkscited by 816
  11. 2018Multi-Task Learning as Multi-Objective Optimizationcited by 809
  12. 2022A Generalist Agentcited by 802
  13. 2017Lifelong Learning with Dynamically Expandable Networkscited by 689
  14. 2017Learning multiple visual domains with residual adapterscited by 553
  15. 2017PathNet: Evolution Channels Gradient Descent in Super Neural Networkscited by 362
  16. 2017Population Based Training of Neural Networkscited by 351
  17. 2019Learn to Grow: A Continual Structure Learning Framework for Overcoming Catastrophic Forgettingcited by 302
  18. 2015Net2Net: Accelerating Learning via Knowledge Transfercited by 245
  19. 2019ERNIE 2.0: A Continual Pre-training Framework for Language Understandingcited by 230
  20. 2018Characterizing and Avoiding Negative Transfercited by 226
  21. 2018Model-Ensemble Trust-Region Policy Optimizationcited by 214
  22. 2021Efficient Large Scale Language Modeling with Mixtures of Expertscited by 175
  23. 2019AdaShare: Learning What To Share For Efficient Deep Multi-Task Learningcited by 159
  24. 2022A Review of Sparse Expert Models in Deep Learningcited by 147
  25. 2022Linearly Mapping from Image to Text Spacecited by 139
  26. 2020A Survey on Negative Transfercited by 120
  27. 2017Incremental Learning Through Deep Adaptationcited by 108
  28. 2020Efficient Continual Learning with Modular Networks and Task-Driven Priorscited by 87
  29. 2022Pathways: Asynchronous Distributed Dataflow for MLcited by 81
  30. 2019Mixture Models for Diverse Machine Translation: Tricks of the Tradecited by 77
  31. 2021Factors of Influence for Transfer Learning across Diverse Appearance Domains and Task Typescited by 54
  32. 2021Model Zoo: A Growing "Brain" That Learns Continuallycited by 53
  33. 2020Online Structured Meta-learningcited by 27
  34. 2022An Evolutionary Approach to Dynamic Introduction of Tasks in Large-scale Multitask Learning Systemscited by 16
  35. 2022muNet: Evolving Pretrained Deep Neural Networks into Scalable Auto-tuning Multitask Systemscited by 12
  36. 2018Evolutionary-Neural Hybrid Agents for Architecture Searchcited by 11
  37. 2022A Continual Development Methodology for Large-scale Multitask Dynamic ML Systemscited by 9
  38. 2019Fast Task-Aware Architecture Inferencecited by 7

50 most cited of 51 All 51 (JSON)

Built on by

Cited by year

Citing papers per year: 2023: 1.

1 paper.

  1. 2023SAFE: Machine Unlearning With Shard Graphscited by 28