mlsanityData to 14 Oct 2026

Learning Where, What and How to Transfer: A Multi-Role Reinforcement Learning Approach for Evolutionary Multitasking

arXiv 2511.15199 ↗First posted 19 Nov 2025cs.NE cs.LGcited by 1

On the map: Deep reinforcement learning methods → Deep RL algorithms and representations

Abstract (arXiv)

Evolutionary multitasking (EMT) algorithms typically require tailored designs for knowledge transfer, in order to assure convergence and optimality in multitask optimization. In this paper, we explore designing a systematic and generalizable knowledge transfer policy through Reinforcement Learning. We first identify three major challenges: determining the task to transfer (where), the knowledge to be transferred (what) and the mechanism for the transfer (how). To address these challenges, we formulate a multi-role RL system where three (groups of) policy networks act as specialized agents: a task routing agent incorporates an attention-based similarity recognition module to determine source-target transfer pairs via attention scores; a knowledge control agent determines the proportion of elite solutions to transfer; and a group of strategy adaptation agents control transfer strength by dynamically controlling hyper-parameters in the underlying EMT framework. Through pre-training all network modules end-to-end over an augmented multitask problem distribution, a generalizable meta-policy is obtained. Comprehensive validation experiments show state-of-the-art performance of our method against representative baselines. Further in-depth analysis not only reveals the rationale behind our proposal but also provide insightful interpretations on what the system have learned.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Deep RL algorithms and representations, one of 15 directions in Deep reinforcement learning methods. Builds on 26 of its 73 references that are in the map. First posted on arXiv 19 Nov 2025.

Builds on

26 of 73 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024Learning to Transfer for Evolutionary Multitasking20 Jun

Other references · 25

  1. 2017Attention Is All You Needcited by 60,361
  2. 2017Proximal Policy Optimization Algorithmscited by 15,215
  3. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 7,005
  4. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
  5. 2017Learning to Generalize: Meta-Learning for Domain Generalizationcited by 763
  6. 2018Deep learning for time series classification: a reviewcited by 576
  7. 2017Inverse Reward Designcited by 249
  8. 2022Multi-agent Dynamic Algorithm Configurationcited by 36
  9. 2023MetaBox: A Benchmark Platform for Meta-Black-Box Optimization with Reinforcement Learningcited by 36
  10. 2020Exploratory Landscape Analysis is Strongly Sensitive to the Sampling Strategycited by 27
  11. 2024Toward Automated Algorithm Design: A Survey and Practical Guide to Meta-Black-Box-Optimizationcited by 24
  12. 2024ConfigX: Modular Configuration for Evolutionary Algorithms via Multitask Reinforcement Learningcited by 15
13 more
  1. 2024Pretrained Optimization Model for Zero-Shot Black Box Optimizationcited by 13
  2. 2021Half a Dozen Real-World Applications of Evolutionary Multitasking, and Morecited by 12
  3. 2025Reinforcement Learning-based Self-adaptive Differential Evolution through Automated Landscape Feature Learningcited by 12
  4. 2017Evolutionary Multitasking for Single-objective Continuous Optimization: Benchmark Problems, Performance Metric, and Baseline Resultscited by 11
  5. 2024Neural Exploratory Landscape Analysis for Meta-Black-Box-Optimizationcited by 10
  6. 2024Deep-ELA: Deep Exploratory Landscape Analysis with Self-Supervised Pretrained Transformers for Single- and Multi-Objective Continuous Optimization Problemscited by 9
  7. 2025MetaDE: Evolving Differential Evolution by Differential Evolutioncited by 9
  8. 2025Meta-Black-Box-Optimization through Offline Q-function Learningcited by 9
  9. 2025MetaBox-v2: A Unified Benchmark Platform for Meta-Black-Box Optimizationcited by 9
  10. 2024RLEMMO: Evolutionary Multimodal Optimization Assisted By Deep Reinforcement Learningcited by 7
  11. 2025Surrogate Learning in Meta-Black-Box Optimization: A Preliminary Studycited by 7
  12. 2025Instance Generation for Meta-Black-Box Optimization through Latent Space Reverse Engineeringcited by 6
  13. 2025DesignX: Human-Competitive Algorithm Designer for Black-Box Optimizationcited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Hyper Algorithm Design Agent: Evolving Learnable Optimizer from Zerocited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023Foundations for Transfer in Reinforcement Learning: A Taxonomy of Knowledge Modalities
  2. 2025InstructRAG: Leveraging Retrieval-Augmented Generation on Instruction Graphs for LLM-Based Task Planning
  3. 2024Task Adaptation of Reinforcement Learning-based NAS Agents through Transfer Learning
  4. 2024AMAGO-2: Breaking the Multi-Task Barrier in Meta-Reinforcement Learning with Transformers
  5. 2025Knowledge Transfer in Model-Based Reinforcement Learning Agents for Efficient Multi-Task Learning
  6. 2026SkillRise: Agentic Reinforcement Learning for Cross-Task Skill Evolution
  7. 2025Multi-Task Multi-Agent Reinforcement Learning via Skill Graphs
  8. 2024A Two-stage Reinforcement Learning-based Approach for Multi-entity Task Allocation