{"678707":{"#nid":"678707","#data":{"type":"event","title":"ML@GT Seminar Series | Deep Learning is Not So Mysterious or Different ","body":[{"value":"\u003Cp\u003EFeaturing Andrew Wilson, New York University\u0026nbsp;\u003C\/p\u003E\u003Cp\u003E\u003Cstrong\u003EAbstract: \u003C\/strong\u003EDeep neural networks are often seen as different from other model classes by defying conventional notions of generalization. Popular examples of anomalous generalization behaviour include benign overfitting, double descent, and the success of overparametrization. We argue that these phenomena are not distinct to neural networks, or particularly mysterious. Moreover, this generalization behaviour can be intuitively understood, and rigorously characterized using long-standing generalization frameworks such as PAC-Bayes and finite hypothesis bounds. We present soft inductive biases as a key unifying principle in explaining these phenomena: rather than restricting the hypothesis space to avoid overfitting, embrace a flexible hypothesis space, with a soft preference for simpler solutions that are consistent with the data. This principle can be encoded in many model classes, and thus deep learning is not as mysterious or different from other model classes as it might seem. However, we also highlight how deep learning is relatively distinct in other ways, such as its ability for representation learning, phenomena such as mode connectivity, and its relative universality.\u003C\/p\u003E\u003Cp\u003E\u003Cstrong\u003EBio: \u003C\/strong\u003EAndrew Gordon Wilson is a Professor at the Courant Institute of Mathematical Sciences and Center for Data Science at New York University. He is interested in developing a prescriptive foundation for\u0026nbsp;\u003C\/p\u003E\u003Cp\u003Ebuilding intelligent systems. His work includes the discovery of mode connectivity, the SWA optimization procedure, the popular GPyTorch library for scalable Gaussian processes, informative generalization bounds for billion parameter neural networks, Bayesian optimization techniques for protein engineering, the first LLM for time-series forecasting, and many contributions to Bayesian deep learning. His website is \u003Ca href=\u0022https:\/\/cims.nyu.edu\/~andrewgw\u0022 target=\u0022_blank\u0022\u003Ehttps:\/\/cims.nyu.edu\/~andrewgw\u003C\/a\u003E.\u003C\/p\u003E","summary":"","format":"limited_html"}],"field_subtitle":"","field_summary":[{"value":"\u003Cp\u003EMachine Learning Center Seminar Series is held bi-weekly on Wednesdays at 12pm.\u0026nbsp;\u003C\/p\u003E","format":"limited_html"}],"field_summary_sentence":[{"value":"Featuring Andrew Wilson, New York University "}],"uid":"36518","created_gmt":"2024-12-02 14:37:51","changed_gmt":"2025-02-14 14:43:15","author":"shatcher8","boilerplate_text":"","field_publication":"","field_article_url":"","field_event_time":{"event_time_start":"2025-02-26T12:00:00-05:00","event_time_end":"2025-02-26T13:00:00-05:00","event_time_end_last":"2025-02-26T13:00:00-05:00","gmt_time_start":"2025-02-26 17:00:00","gmt_time_end":"2025-02-26 18:00:00","gmt_time_end_last":"2025-02-26 18:00:00","rrule":null,"timezone":"America\/New_York"},"location":"CODA 9th Floor Atrium","extras":["free_food"],"hg_media":{"676287":{"id":"676287","type":"image","title":"2025.0226 ML Seminar Announcement-Andrew Wilson.jpg","body":null,"created":"1739544140","gmt_created":"2025-02-14 14:42:20","changed":"1739544140","gmt_changed":"2025-02-14 14:42:20","alt":"ML@GT Seminar Series hosts Andrew Wilson on Wednesday, February 26 at 12pm","file":{"fid":"260041","name":"2025.0226 ML Seminar Announcement-Andrew Wilson.jpg","image_path":"\/sites\/default\/files\/2025\/02\/14\/2025.0226%20ML%20Seminar%20Announcement-Andrew%20Wilson.jpg","image_full_path":"http:\/\/hg.gatech.edu\/\/sites\/default\/files\/2025\/02\/14\/2025.0226%20ML%20Seminar%20Announcement-Andrew%20Wilson.jpg","mime":"image\/jpeg","size":156713,"path_740":"http:\/\/hg.gatech.edu\/sites\/default\/files\/styles\/740xx_scale\/public\/2025\/02\/14\/2025.0226%20ML%20Seminar%20Announcement-Andrew%20Wilson.jpg?itok=2Q5xlSE4"}}},"media_ids":["676287"],"groups":[{"id":"576481","name":"ML@GT"}],"categories":[],"keywords":[{"id":"9167","name":"machine learning"},{"id":"109581","name":"deep learning"}],"core_research_areas":[],"news_room_topics":[],"event_categories":[{"id":"1795","name":"Seminar\/Lecture\/Colloquium"}],"invited_audience":[{"id":"78761","name":"Faculty\/Staff"},{"id":"177814","name":"Postdoc"},{"id":"174045","name":"Graduate students"},{"id":"78751","name":"Undergraduate students"}],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[{"value":"\u003Cp\u003EShelli Hatcher, Program and Operations Manager\u003C\/p\u003E","format":"limited_html"}],"email":[],"slides":[],"orientation":[],"userdata":""}}}