Loading Events

Upcoming Events

Events Search and Views Navigation

Event Views Navigation

October 2019

Charlotta Johnsson — Industry, Innovation and Infrastructure

October 14 @ 11:30 am - 1:00 pm
Hearst Memorial Mining Building 290, Hearst Memorial Mining Building
Berkeley, CA 94720 United States
+ Google Map

Abstract: “The pace of change has never been this fast .. yet it will never again be this slow” (source unknown). This implies that there will be major shifts in our society, including industry and manufacturing. Some even speak about a fourth industrial revolution. We find many names for those we love, hence commonly used terms for this shift include: Smart industry, Smart manufacturing, Industrie 4.0, Industrie du Future, Factory of the Future, Society 5.0, etc. History repeats itself, and…

Find out more »

Ziv Scully – SOAP: One Clean Analysis of All Age-Based Scheduling Policies

October 14 @ 3:30 pm - 4:30 pm
George B. Dantzig Auditorium – 1174 Etcheverry Hall, Etcheverry Hall
Berkeley, CA 94720 United States
+ Google Map

Abstract: Scheduling policies are at the heart of computer systems. The right scheduling policy can dramatically reduce response times, ensure fairness, provide class-based priority, etc., without requiring additional resources. While stochastic response time analysis (e.g. in the M/G/1 queueing model) of different scheduling policies has been the focus of many theoretical papers, results are limited to analyzing a relatively small number of simple scheduling policies, such as first-come, first served (FCFS) and shortest remaining processing time (SRPT), with each requiring…

Find out more »

Karthik Natarajan — Exploiting Partial Correlations in Distributionally Robust Optimization

October 16 @ 3:00 pm - 4:00 pm
3108 Etcheverry Hall

Abstract: In this work, we identify partial correlation information structures that allow for simpler reformulations in evaluating the maximum expected value of mixed integer linear programs with random objective coefficients. To this end, assuming only the knowledge of the mean and the covariance matrix entries restricted to block-diagonal patterns, we develop a reduced semidefinite programming formulation, the complexity of solving which is related to characterizing a suitable projection of the convex hull of the set {(x, xx' ) : x ∈…

Find out more »
+ Export Events