Presentations and/or Posters
Submitters can propose presentations or posters; posters will be displayed during 1.5-hour sessions, and the committee will select 15-minute spotlight presentations.
Machine Learning/CORE Unranked
FastML 2026
Fast Machine Learning for Science Conference 2026
The Fast Machine Learning for Science Conference 2026, hosted by UC San Diego, focuses on emerging machine learning methods and their applications in scientific discovery, with an emphasis on accelerating deep learning and inference. It brings together researchers to share innovations in efficient ML architectures, real-time processing, and scalable systems across domains like physics, genomics, climate modeling, and robotics.
Key deadlines
Verified September 9, 2026Full paper
July 2, 2026
AoE
Conference timeline
Submission and decisions
Full paperKey deadline
July 2, 2026 · AoE
Paper fit
Contribution paths
A strong submission should clearly identify its contribution and evaluate it appropriately.
Tutorials
In-depth 1.5- to 3-hour tutorials on Monday, August 31, with specified levels (beginner/intermediate/advanced).
Topical (Birds-of-a-Feather) Sessions
Participant-driven discussions lasting 1.5 to 3 hours on Thursday, September 3 or Friday, September 4.
Extended Abstracts
4-page papers submitted via OpenReview using the NeurIPS template; references excluded from page limit; accepted contributions will be published on OpenReview and Indico.
Research areas in scope
01
Machine Learning Algorithm Design & Optimization
Novel efficient architecturesHyperparameter optimization and tuningModel compression (quantization, sparsity)Hardware/software co-design for ML efficiencyLarge language model optimization and implementation
02
Accelerated Inference & Real-Time Processing
Low-latency ML for scientific experimentsFPGA/NPU/GPU-based ML accelerationML for trigger systems and data acquisitionOn-detector and edge inference
03
Scientific Applications of Fast ML
High-energy physics, astrophysics, and astronomySpace science and satellite-based MLGenomics and medical imagingClimate and environmental modelingBiological science and neuroscienceFusionQuantum computingMaterial scienceRobotics
04
Scalable & Distributed ML Systems
Cloud-based, accelerated ML processingDistributed inferenceAcceleration-as-a-serviceML compilers and runtimesBenchmarks and datasets
05
Advanced Hardware & Computing Architectures
Specialized AI acceleratorsTools and methodologies for accelerating ML algorithmsHeterogeneous computing platforms for MLBeyond CMOS
Policies worth checking twice
- Submissions do not need to be anonymized; authors may include names, affiliations, acknowledgments, and links to code or data.
- Submissions may include original work, work in progress, or extended-abstract versions of work published or under review elsewhere, provided it fits the conference scope.
- Authors must clearly indicate if the submission is based on previously published or publicly available work.
- Maximum length for extended abstracts is 4 pages, excluding references.
- Format must use the official NeurIPS template in PDF.
- References are unlimited and do not count toward the 4-page limit.
- Appendices are discouraged.
- All authors must be included at the time of submission.
Official sources
Compiled from the official call for papers. The organizers’ pages remain authoritative.
Last verified September 9, 2026