Graduate

Industrial AI presents "Industrial AI and Analytics Business"

Topic: 
Industrial AI and Analytics Business
Abstract / Description: 

Arnab Chakraborty will lead an exciting panel discussion with three very experienced senior executives, who have spent decades in developing real-world industry applications with the use of advanced analytics, data science and machine learning at scale in large corporations. The discussions topics will range from separating the hype from actual results, most common use-cases in their respective industries, current challenges in scaling products and teams, and will also share their thoughts on future vision/trends. The panel will also help students understand the approaches used in industry and how to be prepared to embark on a journey towards a career in this space.

Date and Time: 
Wednesday, April 28, 2021 - 4:00pm

Industrial AI presents "Cloud Infrastructure Analytics"

Topic: 
Cloud Infrastructure Analytics
Abstract / Description: 

Datacenter infrastructure has become ubiquitous worldwide for Cloud computing and many large-scale Internet services. With its rapid growth, reliable and efficient management is key to the success of the business the infrastructure supports. At Alibaba Cloud Intelligence, we focus on using data and the very best techniques that Cloud enables, such as AI algorithms, to manage the Cloud infrastructure itself in an autonomous fashion. In this talk, we give an overview of the top issues large-scale datacenter operation is facing. Then we share some recent progress on specific technologies we've been working on.

Date and Time: 
Wednesday, April 14, 2021 - 4:00pm

Industrial AI presents "From Industrial AI to Systems Intelligence"

Topic: 
From Industrial AI to Systems Intelligence
Abstract / Description: 

Industrial AI is concerned with the application of Artificial Intelligence (AI), Machine Learning (ML) and related technologies towards addressing real-world use cases in industrial and societal domains and will fundamentally transform the world we live in. In this talk, we will first introduce Industrial AI, give some real world example, highlight challenges and lessons learned. We will also show how the current trends in Industrial AI will lead to systematic transformation of our industrial and societal landscape.

Date and Time: 
Wednesday, April 7, 2021 - 4:00pm

Reinforcement Learning Forum Talk: Provable Model-based Nonlinear Bandit and Reinforcement Learning

Topic: 
Provable Model-based Nonlinear Bandit and Reinforcement Learning
Abstract / Description: 

Deep model-based reinforcement learning methods have achieved state-of-the-art sample efficiency but we lack a theoretical understanding of them. This talk will first show that convergence to a global maximum requires an exponential number of samples even for a one-layer neural net bandit problem, which is strictly easier than RL. Therefore, we propose to study convergence to local maxima. For both nonlinear bandit and RL, I will present a model-based algorithm, Virtual Ascent with Online Model Learner (ViOL), which provably converges to a local maximum with sample complexity that only depends on the sequential Rademacher complexity of the model class. Our results imply novel global or local regret bounds on several concrete settings such as linear bandit with finite or sparse model class, and two-layer neural net bandit.

Paper: https://arxiv.org/pdf/2102.04168.pdf


Details available at RL site, link below

 

Date and Time: 
Thursday, April 15, 2021 - 2:30pm

US-ATMC presents "Healthcare Entrepreneurship in China"

Topic: 
Healthcare Entrepreneurship in China
Abstract / Description: 

Over the course of our public seminar series, we will explore the most recent trends, patterns, and challenges of entrepreneurship in Asia and their relevance to Silicon Valley and the U.S. Guest speakers include entrepreneurs, investors and mentors, and other experts on the current entrepreneurial ecosystems of major Asia economies.


Stanford students may register through Axess in course EASTASN 402T (cross listed as EE 402T and EALC 402T) for 1 unit seminar credit.

Date and Time: 
Tuesday, April 13, 2021 - 4:30pm

ISL Colloquium presents "Adaptive Experimental Design for Best Identification and Multiple Testing"

Topic: 
Scalable semidefinite programming
Abstract / Description: 

Semidefinite programming (SDP) is a powerful framework from convex optimization that has striking potential for data science applications. This talk describes a provably correct randomized algorithm for solving large, weakly constrained SDP problems by economizing on the storage and arithmetic costs. Numerical evidence shows that the method is effective for a range of applications, including relaxations of MaxCut, abstract phase retrieval, and quadratic assignment problems. Running on a laptop equivalent, the algorithm can handle SDP instances where the matrix variable has over 10^14 entries.

This talk will highlight the ideas behind the algorithm in a streamlined setting. The insights include a careful problem formulation, design of a bespoke optimization method, use of randomized eigenvalue computations, and use of randomized sketching methods.

Joint work with Alp Yurtsever, Olivier Fercoq, Madeleine Udell, and Volkan Cevher. Based on arXiv 1912.02949 (Scalable SDP, SIMODS 2021) and other papers (SketchyCGM in AISTATS 2017, Nyström sketch in NeurIPS 2017).

Date and Time: 
Thursday, April 8, 2021 - 4:30pm

Pages

Subscribe to RSS - Graduate