The 18th Asian Conference on Machine Learning (ACML 2026) aims to provide a leading international forum for researchers in machine learning and related fields to share their new ideas, progress and achievements. Submissions from regions other than the Asia-Pacific are also highly encouraged.
The conference calls for high-quality, original research papers in the theory and practice of machine learning. The conference also solicits proposals focusing on frontier research, new ideas and paradigms in machine learning. ACML has taken place annually since 2009 in locations throughout the Asia-Pacific region. Previous conferences were held in Taipei, Taiwan (2025), Hanoi, Vietnam (2024), Istanbul, Turkey (2023), Hyderabad, India (2022), Bangkok (converted to virtual), Thailand (2021/2020), Nagoya, Japan (2019), Beijing, China (2018), Seoul, Korea (2017), Hamilton, New Zealand (2016), Hong Kong, China (2015), Nha Trang, Vietnam (2014), Canberra, Australia (2013), Singapore (2012), Taoyuan, Taiwan (2011), Tokyo, Japan (2010), and Nanjing, China (2009).
opics of interest include but are not limited to:
General machine learning
Active learning
Bayesian machine learning
Clustering
Imitation Learning
Learning to Rank
Meta-Learning
Multi-objective learning
Multiple instance learning
Multi-task learning
Neuro-symbolic methods
Online learning
Optimization
Reinforcement learning
Relational learning
Self-supervised learning
Semi-supervised learning
Structured output learning
Supervised learning
Transfer learning
Unsupervised learning
Weakly-supervised learning
Learning with noisy labels
Continual / lifelong learning
Few-shot and zero-shot learning
Out-of-distribution generalization
Other machine learning methodologies
Deep learning
Architectures
Deep reinforcement learning
Generative models
Multi-modality learning
Large-language models and other foundation models
Deep learning theory
Representation learning
Other topics in deep learning
Generative AI
Multimodal generative models
Controllable and conditional generation
Editing, inpainting, and style transfer
Evaluation
Creative applications (art, music, media)
Theory
Bandits
Computational learning theory
Game theory
Optimization
Statistical learning theory
Other theories
Datasets and reproducibility
Implementations, libraries
ML datasets and benchmarks
Synthetic data generation
Other topics in reproducible ML research
Trustworthy machine learning
Accountability, explainability, transparency
Adversarial learning
Causality
Fairness
Privacy
Robustness
AutoML
AI safety and alignment
Hallucination mitigation and reliability
Other topics in trustworthy ML
Learning in knowledge-intensive systems
Knowledge refinement and theory revision
Multi-strategy learning
Retrieval-augmented generation (RAG)
Knowledge-enhanced foundation models
Other systems
Applications
Bioinformatics
Biomedical informatics
Climate science
Collaborative filtering
Computer vision
Healthcare
Human activity recognition
Information retrieval
Natural language processing
Social good
Social networks
Web search
ML for science discovery
Other applications
Similar to previous years, ACML 2026 offers two publication tracks: the conference track and the journal track. Please note that at least one author of each accepted paper (for both tracks) must present the paper at the conference. Failure to do so will result in the paper not being published. All deadlines will be at 23:59 AoE (Anywhere on Earth) unless otherwise specified.
To maintain high-quality peer review and support our growing community, ACML 2026 encourages all authors to serve as a reviewer. At least one author must be nominated for each submission, unless one author has served as an area chair. Author(s) with extensive research and reviewing experiences at top-tiered machine learning venues (e.g., ICML, NeurIPS, ICLR) are highly preferred to be nominated. Submissions without at least one author serving as a reviewer or area chair will be desk rejected. If your submission cannot nominate any authors that satisfy the above requirement, please email the PCs before 23 June 2026 through the following email: acml_2026_programchairs@googlegroups.com. You need to include the OpenReview submission ID in the email so we can process your information.
Conference TrackPermalink
Conference Track: (16-page limit with references) for which the proceedings will be published as a volume of Proceedings of Machine Learning Research Workshop and Conference Proceedings (PMLR).
Submission Deadline: 26 June 2026
For the conference track, please submit your manuscript via OpenReview at: https://openreview.net/group?id=ACML.org/2026/Conference. Manuscripts must be written in English, and should follow the Latex submission template and style file here ACML_camera_ready.zip with a 16-page limit, including references and appendix. Supplementary materials may be submitted as a separate file, but reviewers are not obliged to consider it.
All conference track submissions must be anonymized for double-blinded review. Submissions that are not anonymized, over-length, or not in the correct format will be rejected without review. To anonymize, simply leave the author information empty in the Tex template. There is no separate format for anonymizing.
It is not appropriate to submit papers that are substantially similar to versions that have been previously published, or accepted for publication, or that have been submitted in parallel to other conferences or journals (including our journal track). However, submission is permitted for papers presented or to be presented at conferences or workshops without proceedings, or with only abstracts published. Also, submission is permitted for papers that are available as a technical report (e.g., in arXiv) as long as it is not cited in the submission.
Journal TrackPermalink
Submission Deadline: 20 June 2026
In addition to the conference track, this year’s ACML will also run a journal track, similar to previous years. Papers that are accepted to the journal track must be presented at the conference in order to be published.
IMPORTANT: Similar to previous years, for the journal track, the abstract and the paper must be submitted to two different systems simultaneously for the purpose of review management:
1) First, please submit the title, abstract, and the full manuscript via OpenReview at ACML 2026 Journal Track | OpenReview.
2) Then, please submit the full manuscript via Springer Nature’s manuscript submission system at: ACML 2026 | SpringerLink. When creating a new submission, please make sure to choose “Research” as the article type and “ACML 2026” as the collection type.
Failure to submit to both systems will result in desk-reject of the paper.
For the journal track, manuscripts must be written in English with a maximum of 20 pages (including references, appendices, etc.). For the template and style files, please follow the submission guidelines on the journal website.
The journal track will follow the reviewing process of the Machine Learning journal. This includes allowing papers that require minor changes to be resubmitted after a first-round review. The journal track committee will aim to complete the reviewing process in time for this year’s conference. In the unlikely event that the reviewing process for a paper is not completed in time (for this year’s conference), the paper will not be considered for the conference and the review will be completed as a regular submission to the Machine Learning journal.
The journal track review is single-blind, i.e., the authors’ identity will be visible to reviewers. It is not appropriate to submit papers that are substantially similar to versions that have been previously published, or accepted for publication, or that have been submitted in parallel to other conferences or journals. Submissions that are not in the correct format will be rejected without review. In addition, extended versions of published conference papers are not eligible for journal track submission. However, submission is permitted for papers presented or to be presented at conferences or workshops without proceedings, or with only abstracts published. Also, submission is permitted for papers that are available as a technical report (e.g., in arXiv).