[Cs-affiliates] Remaining department visits - Wednesday 3/29 (today), Monday April 3, Friday April 7

Devits, Christopher R cdevits at bu.edu
Wed Mar 29 09:23:27 EDT 2017

Good Wednesday Morning Everyone,

A reminder that today we have Emily Whiting visiting us. Her talk will be at 11:00AM in Hariri and the graduate student meeting will be at 3:15-4:00PM, location Room 148. Her talk details are below.

Additionally, we have two late additions to the schedule for this semester. Next Monday (April 3) and Friday (April 7) we will have guests visiting us. Monday, we will host Brad Hayes, and Friday we will host Ron Rothblum. Both their talks will be at 11:00AM in the Hariri Institute, and student meetings are TBD. Their talk information is below as well.

As always, thanks for your participation this semester.



Presenter: Emily Whiting
Seminar Date + time: Wednesday March 29, 11:00AM
Location: Hariri Institute Seminar Room

Abstract: With the advent of rapid prototyping technologies such as desktop 3D printers, it is now simple to produce physical realizations of complex 3D models. Yet today's tools for creating digital content are largely unaware of the fundamental laws that govern how structures behave in the real world. 3D modeling software typically gives no indication of gravity, support or other properties of mechanics. My research aims to establish a field of Mechanics-Aware Geometry Processing: a cross-pollination of digital geometry processing and rapid prototyping, leveraging techniques in constrained optimization, numerical methods, and geometric computing. For example, our recent work in architectural geometry reduces labor and material in assembly of self-supporting block structures, where optimal construction sequences are found by solving a combinatorial problem. On the same theme, we developed techniques that optimize the internal structure of 3D printed objects to achieve design goals including static equilibrium, rotational inertia, buoyancy, and acoustics of wind instruments. This work has been featured in numerous media sources including TEDx, MIT Technology Review, and Make magazine.

Website: http://www.cs.dartmouth.edu/~emily/

Presenter: Brad Hayes
Seminar Date + time: Monday April 3, 11:00AM
Location: Hariri Institute Seminar Room

Title: Learning to be a Good Teammate -- Algorithms for Effective Human-Robot Collaboration.

Abstract: Robots capable of fluent collaboration with humans will bring transformative changes to the way we live and work. In domains ranging from healthcare to education to manufacturing, particularly under conditions where modern automation is ineffective or inapplicable, human-robot teaming can be leveraged to increase efficiency, capability, and safety. Despite this, the deployment of collaborative robots into human-dominated environments remains largely infeasible due to the myriad challenges involved in creating helpful, safe, autonomous teammates.

In this talk I will present my recent work in overcoming these challenges, toward realizing flexible, communicative robot collaborators that both learn and dynamically assist in the completion of complex tasks through the application of novel learning and control algorithms. In particular, I will cover approaches to hierarchical task modeling, task and motion planning, and cooperative inverse reinforcement learning within the theme of human-robot teaming, with a particular focus on recently published work concerning the autonomous synthesis of natural language explanations that allow robots to describe their control policies to humans.

Bio: Dr. Bradley Hayes is a Postdoctoral Associate in the Interactive Robotics Group within the Computer Science and Artificial Intelligence Laboratory at the Massachusetts Institute of Technology, working with Professor Julie Shah. Brad's research interests center around developing the algorithms necessary for building supportive, interactive, and intuitive robotic systems that are capable of performing complex collaborative tasks in environments shared with humans. His work combines theoretical advances and practical applications of machine learning, task and motion planning, human teaming psychology, and human-robot interaction.

Presenter: Ron Rothblum
Seminar Date + time: Friday April 7, 11:00AM
Location: Hariri Institute Seminar Room

Title: How to Prove the Correctness of Computations.

Abstract: In recent years much of the focus in cryptography has shifted from the classical goals of securing communication to securing computation. Authenticating the correctness of computations has emerged as a central challenge in this new frontier. For example, we want to enable a client with limited computational resources (such as a smartphone or tablet) to outsource an expensive computation to a powerful, but untrusted, server.

In the talk we will discuss new results that allow the client to verify the correctness of the computation. These techniques, which are generic (rather than being tailored to specific computations), ensure that:

1. The client's verification process is extremely efficient (and in particular, far less expensive than re-executing the computation).

2. The server's overhead in proving the correctness of the computation is minimal (i.e., not much more than the cost of merely performing the computation).

Website: http://people.csail.mit.edu/ronr/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://cs-mailman.bu.edu/pipermail/cs-affiliates/attachments/20170329/cfb1e174/attachment-0001.html>

More information about the Cs-affiliates mailing list