AISTATS 2027 — International Conference on Artificial Intelligence and Statistics

AISTATS 2027, the International Conference on Artificial Intelligence and Statistics, takes place 3–6 May 2027 in Montreal, Canada. AISTATS describes itself as "an interdisciplinary gathering of researchers from computer science, artificial intelligence, machine learning, statistics, and related areas" whose mission is to "broaden research in these fields by promoting the exchange of ideas among them". Its call covers machine learning methods and algorithms; probabilistic methods including Bayesian inference, density estimation and probabilistic programming; and the theory of machine learning and statistics — optimization, computational learning theory, decision theory, online learning and bandits, game theory and information theory — alongside deep learning, reinforcement learning, ethical AI and applications in NLP and computer vision. Papers run to eight pages and are selected by double-blind peer review. Abstracts closed 29 September 2026, full papers 6 October 2026, reviews are released 16 November 2026 and decisions land 20 January 2027.

• Abstract submission deadlineSep 29, 2026
• Full paper submission deadlineOct 6, 2026
• Reviews released to authorsNov 16, 2026
• Notification to authorsJan 20, 2027

Date

May 3–6, 2027

Location

Montreal, Canada

Format

In-Person

Event Type

Conference

Status

Upcoming

Pricing

Paid

Organizer

Artificial Intelligence and Statistics (AISTATS)

Added

Sep 7, 2026

Updated

Sep 7, 2026

• ✓Machine learning researchers
• ✓Statisticians and data scientists
• ✓Applied scientists in industrial research labs
• ✓Heads of data science
• ✓Quantitative researchers in finance and healthcare
• ✓AI/ML PhD students

For a VP of data science or a head of AI research, AISTATS is where the methods that later become production tooling are argued rigorously first — calibrated uncertainty, approximate Bayesian inference, bandit-driven experimentation, online learning under constraints. It is a concentrated recruiting pool for the statistically trained researchers that large-model teams are short of, and its early-May slot means the work presented is what shows up in the summer submission cycle six months later. Montreal keeps travel and accommodation costs well below the equivalent US venue for North American and European teams.

• ✓Main technical programme selected by "a rigorous double-blind peer-review process"
• ✓Topic area: machine learning methods and algorithms — classification, regression, unsupervised and semi-supervised learning, clustering, logic programming
• ✓Topic area: probabilistic methods — Bayesian methods, approximate inference, density estimation, tractable probabilistic models, probabilistic programming
• ✓Topic area: theory of machine learning and statistics — optimization, computational learning theory, decision theory, online learning and bandits, game theory, information theory
• ✓Submissions welcomed on deep learning, reinforcement learning, ethical AI and applications in natural language processing and computer vision
• ✓Eight-page paper limit excluding references, the reproducibility checklist and appendices

• ★One of the few major venues that treats statistics and machine learning as a single field — its stated mission is to "broaden research in these fields by promoting the exchange of ideas among them".
• ★Probabilistic and Bayesian methods, approximate inference and probabilistic programming are first-class topic areas rather than a niche track.
• ★An eight-page limit and double-blind review keep the programme tighter than the large general machine learning conferences.
• ★The full review timeline is published more than a year ahead — reviews released 16 November 2026, decisions 20 January 2027 — which is unusual scheduling certainty for authors and travel planners.

Registration pricing for the 2027 edition has not been published as of September 2026 — the official site carries the dates, location and the full submission timeline but not the registration or sponsor portal, which it says will be announced as they are confirmed.

When is AISTATS 2027?

AISTATS 2027 takes place May 3–6, 2027 in Montreal, Canada.

Where is AISTATS 2027 held?

AISTATS 2027 is held in Montreal, Canada. The event format is in-person.

What is the AISTATS 2027 submission deadline?

The next AISTATS 2027 deadline is abstract submission deadline on Sep 29, 2026. Remaining dates: Full paper submission deadline — Oct 6, 2026; Reviews released to authors — Nov 16, 2026; Notification to authors — Jan 20, 2027.

How much does AISTATS 2027 cost?

AISTATS 2027 is paid. Registration pricing for the 2027 edition has not been published as of September 2026 — the official site carries the dates, location and the full submission timeline but not the registration or sponsor portal, which it says will be announced as they are confirmed.

Who should attend AISTATS 2027?

AISTATS 2027 is aimed at Machine learning researchers, Statisticians and data scientists, Applied scientists in industrial research labs, Heads of data science, Quantitative researchers in finance and healthcare, AI/ML PhD students.

What topics does AISTATS 2027 cover?

AISTATS 2027 covers Machine learning methods and algorithms, Probabilistic methods, Bayesian inference, Learning theory, Optimization, Online learning and bandits, Deep learning, Reinforcement learning, Ethical AI.

This page was written from 4 sources, 2 on domains other than virtual.aistats.org.

1. 1.virtual.aistats.org — virtual.aistats.orgvendor
2. 2.virtual.aistats.org — CallForPapersvendor
3. 3.opencurious.com — ai conference deadlines
4. 4.underleaf.ai — conference deadlines