Events

Nov13

Why Autonomy is Hard

HP Auditorium, 306 Soda Hall
  • Dr. Brandon Basso, Director of Autonomy at Uber Advanced Technologies Group (ATG), UBER
Dr Brandon Basso is a former UC Berkeley graduate from the ME department and the Director of Autonomy at Uber Advanced Technologies Group (ATG). In this talk he will give an overview of the autonomous car program at UBER and focus on the challenges associated to delivering safe autonomous cars. BIOGRAPHY Brandon Basso is a Director of Autonomy at Uber Advanced Technologies Group...
Nov13

Integrating eco-evolutionary data from islands to infer biodiversity dynamics: Berkeley Distinguished Lectures in Data Science

190 Doe Library
  • Rosemary Gillespie, Professor, Environmental Science, Policy, and Management, UC Berkeley
A central challenge in understanding the origins of biodiversity is that, while we can observe and test local ecological phenomena, we must usually infer the longer-term outcomes of these ecological forces indirectly. My colleagues and I have been developing inferential models at the interface between macroecology and population-level processes, and applying them to data from geological...
Nov14

Scientific Computing and Matrix Computations Seminar: Estimating a manifold from noisy samples

380 Soda Hall
  • Yariv Aizenbud, Tel Aviv Univ.
Estimating a manifold from (possibly noisy) samples appears to be a difficult problem. Indeed, even after decades of research, all manifold learning methods do not actually "learn" the manifold, but rather try to embed it into a low-dimensional Euclidean space. This process inevitably introduces distortions and cannot guarantee a robust estimate of the manifold. In this talk, we will discuss a...
Nov14

Scientific Computing and Matrix Computations Seminar: Estimating a manifold from noisy samples

380 Soda Hall
  • Yariv Aizenbud, Tel Aviv Univ.
Estimating a manifold from (possibly noisy) samples appears to be a difficult problem. Indeed, even after decades of research, all manifold learning methods do not actually "learn" the manifold, but rather try to embed it into a low-dimensional Euclidean space. This process inevitably introduces distortions and cannot guarantee a robust estimate of the manifold. In this talk, we will discuss a...
Nov14

Unimodular uniformization and random walks

1011 Evans Hall
  • James R. Lee, University of Washington
Consider deforming the path metric of a unimodular random graph by a (unimodular) reweighting of its vertices. In many instances, a well-chosen change of metric can be used to study the spectral measure, estimate the heat kernel, and bound the speed of the random walk. Even for extensively studied models like random planar maps (e.g., the uniform infinite planar triangulation) and critical...
Nov14

Berkeley ACM A.M. Turing Laureate Lecture: Game Theory in Auction and Blockchain with Andrew Yao

Banatao Auditorium Sutardja Dai Hall
  • Andrew Chi-Chih Yao, Tsinghua Univeristy, Beijing
Game theory has provided a mathematical framework for addressing issues in many domains, including economics and distributed computing systems. In recent years the rise of novel commercial infrastructure, such as electronic auction (for on-line ads) and blockchain, has led to many new interdisciplinary studies involving algorithmic games. In this talk we discuss some recent work from this...
Nov14

VIDEO GAME Playtest!

241 Cory Hall
The Nuclear Science and Security Consortium (NSSC) and the Project on Nuclear Gaming (PoNG) invite you to playtest a game called Signal, developed in partnership between UC Berkeley students and scientists from both Sandia and Lawrence Livermore National Laboratories. Signal will be playtested in both a VIDEO GAME and BOARD GAME format. This game will be used as part of a research effort toward...
Nov14
Logistic regression is one of the most popular methods in binary classification, wherein estimation of model parameters is carried out by solving the maximum likelihood (ML) optimization problem, and the ML estimator is defined to be the optimal solution of this problem. It is well known that the ML estimator exists when the data is non-separable, but fails to exist when the data is separable....
Nov14

Evening with Industry

Berkeley City Club
Join Berkeley SWE and various large engineering firms at Evening with Industry for an evening of networking and fine dining at Berkeley City Club on Wednesday, November 14! This year, SWE is inviting representatives from Google, Intel, IBM, AT&T, P&G, Dropbox, General Motors, KLA Tencor, Danaher, Accenture Labs, PWC, AECOM, Intuitive Surgical, STRIVR, and ZS Associates. If you’re currently...