*Result*: RL-I2IT: Image-to-image translation with deep reinforcement learning.

Title:
RL-I2IT: Image-to-image translation with deep reinforcement learning.
Authors:
Hu J; School of Computer Science, Chengdu University of Information Technology, Chengdu, Sichuan, 610225, China., Luo Z; Department of Information Technology, Uppsala University, Uppsala, Uppsala, 75105, Sweden., Feng C; School of Computer Science, Chengdu University of Information Technology, Chengdu, Sichuan, 610225, China., Hu S; School of Applied and Creative Computing, Purdue University, West Lafayette, IN, 46202, USA., Zhu B; Microsoft Research Asia, No. 5 Danling Street, Tower 1, First Floor, Haidian District, Beijing, Beijing, 100080, China., Wu X; School of Computer Science, Chengdu University of Information Technology, Chengdu, Sichuan, 610225, China., Li X; Department of Computer Science, University at Albany (SUNY), NY, NY, 12222, USA., Zhu H; Departments of Biostatistics, Statistics, Computer Science, and Genetics, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, 27514, USA., Lyu S; Department of Computer Science and Engineering, University at Buffalo (SUNY), NY, NY, 12222, USA., Wang X; College of Integrated Health Sciences, and AI Plus Institut, University at Albany (SUNY), NY, NY, 12222, USA. Electronic address: xwang56@albany.edu.
Source:
Neural networks : the official journal of the International Neural Network Society [Neural Netw] 2026 Mar; Vol. 195, pp. 108264. Date of Electronic Publication: 2025 Oct 29.
Publication Type:
Journal Article
Language:
English
Journal Info:
Publisher: Pergamon Press Country of Publication: United States NLM ID: 8805018 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1879-2782 (Electronic) Linking ISSN: 08936080 NLM ISO Abbreviation: Neural Netw Subsets: MEDLINE
Imprint Name(s):
Original Publication: New York : Pergamon Press, [c1988-
Contributed Indexing:
Keywords: Deep learning; Deep reinforcement learning; Generative model; Image to image translation; Meta policy
Entry Date(s):
Date Created: 20251111 Date Completed: 20260124 Latest Revision: 20260128
Update Code:
20260130
DOI:
10.1016/j.neunet.2025.108264
PMID:
41218403
Database:
MEDLINE

*Further Information*

*Most existing Image-to-Image Translation (I2IT) methods generate images in a single run of deep learning (DL) models. However, designing a single-step model often requires many parameters and suffers from overfitting. Inspired by the analogy between diffusion models and reinforcement learning, we reformulate I2IT as an iterative decision-making problem via deep reinforcement learning (DRL) and propose a computationally efficient RL-based I2IT (RL-I2IT) framework. The key feature in the RL-I2IT framework is to decompose a monolithic learning process into small steps with a lightweight model to progressively transform the source image to the target image. Considering the challenge of handling high-dimensional continuous state and action spaces in the conventional RL framework, we introduce meta policy with a new "concept Plan" to the standard Actor-Critic model. This plan is of a lower dimension than the original image, which facilitates the actor to generate a tractable high-dimensional action. In the RL-I2IT framework, we also employ a task-specific auxiliary learning strategy to stabilize the training process and improve the performance of the corresponding task. Experiments on several I2IT tasks demonstrate the effectiveness and robustness of the proposed method when facing high-dimensional continuous action space problems. Our implementation of the RL-I2IT framework is available at https://github.com/lesley222/RL-I2IT.
(Copyright © 2025. Published by Elsevier Ltd.)*

*Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.*