On Gradient-Based Optimization: Accelerated, Stochastic and Nonconvex

Channel:
Microsoft Research
Subscribers:
351,000
Published on ● Video Link: https://www.youtube.com/watch?v=ft9Abw6VULo


Duration: 1:07:08
5,263 views
82

Many new theoretical challenges have arisen in the area of gradient-based optimization for large-scale statistical data analysis, driven by the needs of applications and the opportunities provided by new hardware and software platforms. I discuss several recent, related results in this area: (1) a new framework for understanding Nesterov acceleration, obtained by taking a continuous-time, Lagrangian/Hamiltonian/symplectic perspective, (2) a discussion of how to escape saddle points efficiently in nonconvex optimization, and (3) the acceleration of Langevin diffusion.

See more at https://www.microsoft.com/en-us/research/videos/ai-distinguished-lecture-series/



Other Videos By Microsoft Research


2018-05-29AI and Our Future With Machines with Dr. Eric Horvitz
2018-05-29Snippets from the Revolution – An Interview with Dr. Jaime Teevan
2018-05-29PNW PLSE Workshop: Helping Designers Explore the Space of Layout Variations with Constraints
2018-05-29NW-NLP 2018: Annotation Artifacts in Natural Language Inference Data
2018-05-29Unsupervised Discovery of Objects and their Interactions for Common-Sense Physical Reasoning
2018-05-29NW-NLP 2018: Ben Taskar Invited Talk; Learning and Reasoning about the World using Language
2018-05-29Towards Autonomous AI-based Measurement Systems
2018-05-29NW-NLP 2018: Compositional Language Modeling for Icon-Based Augmentative & Alternative Communication
2018-05-28Matching and Dynamic Pricing in Ride-Hailing Platforms
2018-05-28Cryptographic Program Obfuscation: Current Capabilities and Challenges
2018-05-28On Gradient-Based Optimization: Accelerated, Stochastic and Nonconvex
2018-05-28NW-NLP 2018: Adverbial Clausal Modifiers in the LinGO Grammar Matrix
2018-05-23Advancing Accessibility with Dr. Meredith Ringel Morris
2018-05-18ISCA 2018 Lightning Talk: Project Brainwave
2018-05-16Not Lost in Translation with Arul Menezes
2018-05-15Clouds, catapults and life after the end of Moore’s Law with Dr. Doug Burger
2018-05-15Machine learning and the incredible flying robot with Dr. Ashish Kapoor
2018-05-15Brokering Peace Talks in the Networking and Storage Arms Race with Dr. Anirudh Badam
2018-05-15Tiny Functions for Codecs, Compilation, and (maybe) soon Everything
2018-05-15Sparsity-aware Sound Field Capturing / Joint Source and Sensor Placement for Sound Field Control
2018-05-14Quantum Computing for Computer Scientists


Tags:
microsoft research