The Conference on Neural Information Processing Systems (NeurIPS, formerly NIPS) is an annual academic conference held in December, focusing on machine learning and computational neuroscience. It is recognized as one of the three primary high-impact conferences in machine learning and artificial intelligence research, alongside the International Conference on Learning Representations (ICLR) and the International Conference on Machine Learning (ICML). The conference features three days of invited talks, oral and poster presentations of refereed papers, followed by two days of workshops and competitions.
History
The NeurIPS meeting was first proposed in 1986 at the annual invitation-only Snowbird Meeting on Neural Networks for Computing, organized by the California Institute of Technology and Bell Laboratories. NeurIPS was designed as a complementary open interdisciplinary meeting for researchers exploring biological and artificial neural networks. The first conference took place in 1987, with information theorist Ed Posner as conference president and learning theorist Yaser Abu-Mostafa as program chairman. Early meetings covered a wide range of topics, from engineering problems to using computer models to understand biological nervous systems. Over time, the biological and artificial systems research streams diverged, and recent proceedings are dominated by papers on machine learning, artificial intelligence, and statistics.
From 1987 to 2000, NeurIPS was held in Denver, United States. Subsequent locations included Vancouver, Canada (2001–2010), Granada, Spain (2011), Lake Tahoe, United States (2012–2013), Montreal, Canada (2014–2015), Barcelona, Spain (2016), Long Beach, United States (2017), Montreal again (2018), and Vancouver in 2019. The 2020 and 2021 editions were virtual due to the COVID-19 pandemic. The conference returned to in-person events in New Orleans in 2022 and 2023, then Vancouver in 2024. Future locations include San Diego and Mexico City in 2025, and Sydney in 2026 with satellite events in Atlanta and Paris.
The first NeurIPS Conference was sponsored by the IEEE. Subsequent conferences have been organized by the NeurIPS Foundation, established by Ed Posner. Terrence Sejnowski has been president of the foundation since Posner's death in 1993. The board of trustees consists of previous general chairs of the conference. The first proceedings were published in book form by the American Institute of Physics in 1987, titled Neural Information Processing Systems. Subsequent proceedings have been published by Morgan Kaufmann (1988–1993), MIT Press (1994–2004), and Curran Associates (2005–present) under the name Advances in Neural Information Processing Systems.
The conference was originally abbreviated as "NIPS." By 2018, some commentators criticized the abbreviation for encouraging sexism due to its association with the word "nipples" and for being a slur against Japanese. The board changed the abbreviation to "NeurIPS" in November 2018.
Topics
Along with machine learning and neuroscience, other fields represented at NeurIPS include cognitive science, psychology, computer vision, statistical linguistics, and information theory. Over the years, NeurIPS became a premier conference on machine learning. Although the "Neural" in the acronym had become somewhat historical, the resurgence of deep learning since 2012, fueled by faster computers and big data, led to achievements in speech recognition, object recognition, image captioning, language translation, and world championship performance in the game of Go. These advances were based on neural architectures inspired by the visual cortex (convolutional networks) and reinforcement learning inspired by the basal ganglia (temporal difference learning).
Notable affinity groups have emerged from NeurIPS, including Black in AI (2017) and Queer in AI (2016), reflecting a commitment to diversity.
Named Lectures
NeurIPS organizes two named lectureships to recognize distinguished researchers. The Posner Lectureship, introduced in honor of founder Ed Posner, was given twice a year up to 2015. Past Posner lecturers include Josh Tenenbaum and Michael I. Jordan (2010), Thomas Dietterich and Terry Sejnowski (2012), Daphne Koller and Peter Dayan (2013), Yann LeCun (2016), Yoshua Bengio (2019), and Christopher Bishop (2020). The Breiman Lectureship, introduced in 2015, highlights work in statistics relevant to conference topics, named for statistician Leo Breiman. Past Breiman lecturers include Robert Tibshirani (2015), Yee Whye Teh (2017), Bin Yu (2019), and Arnaud Doucet (2024).
Review Consistency Experiment
At NIPS 2014, program chairs duplicated 10% of all submissions and sent them through separate reviewers to evaluate randomness in the reviewing process. The results showed that 60% of duplicated submissions received the same accept/reject decision, leading researcher John Langford to note that while the decision is not completely random (a purely random process would have ~78% arbitrariness), the 60% figure is much closer to 78% than 0%, indicating that the outcome is mostly arbitrary. In NeurIPS 2021, the experiment was repeated, finding similar inconsistency: 23% of duplicated submissions received different decisions, and 50.6% of accepted papers would have been rejected under re-review.