International Conference on Learning Representations

The International Conference on Learning Representations (ICLR) is widely considered one of the three most prestigious and influential machine learning conferences in the world, alongside NeurIPS and ICML. It was founded in 2012 and first held in May 2013 in Scottsdale, Arizona, by Yann LeCun (New York University, later Meta AI) and Yoshua Bengio (Université de Montréal), both of whom are Turing Award recipients recognized for their foundational contributions to deep learning. ICLR was established as an experimental venue to explore open peer review and the rapid dissemination of cutting-edge ideas in representation learning — the study of how to automatically learn useful features and representations from data. This focus on representation learning broadly encompasses what is now commonly called deep learning. Since its inception, ICLR has used an open review system (via OpenReview.net) in which submitted papers, reviews, and author responses are publicly visible, a practice that was novel in academic ML publishing at the time. The conference is held annually and rotates locations internationally. Recent and upcoming editions include ICLR 2025 in Singapore and ICLR 2026 scheduled for April 23-27, 2026 at the Riocentro Convention Center in Rio de Janeiro, Brazil. ICLR is governed by a board and organized by a rotating program and organizing committee of volunteer researchers. The conference is funded primarily through corporate sponsorships from major technology companies including Google, Microsoft, Meta, Amazon, Apple, and others. As an academic venue rather than a standalone nonprofit organization, ICLR does not have a traditional organizational structure with full-time staff or published operating budgets. It serves as the primary venue where much of the world's most impactful deep learning research is first presented and disseminated to the broader research community.