Select a calendar:
Filter September Events by Event Type:
SUNMONTUEWEDTHUFRISAT
Events for September 05, 2013
-
PhD Defense - Jun-young Kwak
Thu, Sep 05, 2013 @ 09:00 AM - 11:00 PM
Thomas Lord Department of Computer Science
University Calendar
PhD Candidate: Jun-young Kwak
Committee members:
Milind Tambe (chair)
Rajiv Maheswaran
Yu-Han Chang
Burcin Becerik-Gerber
Wendy Wood (outside member)
Pradeep Varakantham
Time: September 5th 9am-11am
Location: RTH 526
Title: The Power of Flexibility: Towards Agents That Conserve Energy in Commercial Buildings
Abstract:
Agent-based systems for energy conservation are now a growing area of research in multiagent systems, with applications ranging from energy management and control on the smart grid, to energy conservation in residential buildings, to energy generation and dynamic negotiations in distributed rural communities. Contributing to this area, my thesis presents new agent-based models and algorithms aiming to conserve energy in commercial buildings.
More specifically, my thesis provides three sets of algorithmic contributions. First, when multiple users contribute to energy savings, fair division of credit for such savings arises as an important question. I appeal to cooperative game theory and specifically to the concept of Shapley value for this fair division. Unfortunately, scaling up this Shapley value computation is a major hindrance in practice. Therefore, I present novel approximation algorithms to efficiently compute the Shapley value based on sampling and partitions and to speed up the characteristic function computation. Second, I present a novel BM-MDP (Bounded-parameter Multi-objective Markov Decision Problem) model and robust algorithms for multi-objective optimization under uncertainty both at the planning and execution time. The BM-MDP model and its robust algorithms are useful in (re)scheduling events to achieve energy efficiency in the presence of uncertainty over user's preferences. Third, I provide online predictive scheduling algorithms to handle massive numbers of meeting/event scheduling requests considering flexibility, which is a novel concept for capturing generic user constraints while optimizing the desired objective.
These new models have not only advanced the state of the art in multiagent algorithms, but have actually been successfully integrated within two agents dedicated to energy efficiency: SAVES and TESLA. SAVES focuses on the day-to-day energy consumption of individuals and groups in commercial buildings by reactively suggesting energy conserving alternatives. TESLA takes a long-range planning perspective and optimizes overall energy consumption of a large number of group events or meetings together. While SAVES and TESLA thus differ in their scope and applicability, both demonstrate the utility of agent-based systems in actually reducing energy consumption in commercial buildings.
I evaluate my algorithms and agents using extensive analysis on data from over 110,000 real meetings/events at multiple educational buildings including the main libraries at the University of Southern California. I also provide results on simulations and real-world experiments, clearly demonstrating the power of agent technology to assist human users in saving energy in commercial buildings.
Location: Ronald Tutor Hall of Engineering (RTH) - 526
Audiences: Everyone Is Invited
Contact: Lizsl De Leon
-
CS Colloquium Series: Dr Pradeep Varakantham: Multi-Agent Systems for improving Quality of Life in Urban Environments
Thu, Sep 05, 2013 @ 12:00 PM - 01:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Dr Pradeep Varakantham, Singapore Management University
Talk Title: Multi-Agent Systems for improving Quality of Life in Urban Environments
Series: CS Colloquium
Abstract: In this talk, I will present our research on large scale multi-agent systems for improving quality of life in urban cities of today. Technically, we focus on problems of allocating resources to multiple agents in cooperative, selfish or adversarial settings, while considering different objectives, e.g., maximizing revenue or utility, minimizing energy consumption or wait times, etc. We have provided generic solutions to these problems that are at the intersection of Artificial Intelligence (Planning and Scheduling), Game Theory, Behavioral Economics and Optimization. Finally, we have demonstrated the utility of our techniques in the context of applications in:
(a) Transportation: By developing extensions to the well known Congestion Games model to account for involuntary movements of taxi drivers (dictated by customer movement) and providing scalable mechanisms for solving the new representation, we optimized taxi fleet operations of a major taxi company (more than 8000 taxis) with respect to revenue of taxi drivers and availability of taxis.
(b) Leisure/Entertainment: By exploiting network structure and limited impact of each individual patron's movement, our work builds on reward sharing games and orienteering problems to minimize wait times for individual patrons at large theme parks. This was demonstrated on a well known theme park in Singapore.
(c) Energy: In conserving energy at office buildings, we proposed new approaches for scheduling meetings that are based on exploiting flexibility of individual participants. By analyzing 32k meeting requests, studying user behaviors w.r.t providing flexibility in meeting requests and exploiting the flexibility, we predicted a potential benefit of 17k$ annually at one of the buildings in University of Southern California with our approach.
(d) Security: Based on using Stackelberg Games, we have developed approaches to compute randomized patrolling strategies for protecting the rail networks in many large cities of today. This was demonstrated on Singapore rail network that consists of more than 100 stations spread over 7 lines.
Biography: Pradeep Varakantham received his Ph.D. degree in Computer Science from the University of Southern California and he was a post doctoral fellow at Carnegie Mellon University. Currently, he serves as assistant professor at Singapore Management University, where he teaches advanced topics in intelligent decision support, which includes techniques on distributed problem solving; planning/scheduling; and game-theoretic approaches. He is author or co-author of more than 40 international publications and has served as co-chair of the International Workshop on Multiagent Sequential Decision Making under Uncertainty in 2007 and 2008, and also the AAAI Symposium on Multi-Agent Coordination under Uncertainty in 2011. He has also served on the program committee of most major conferences (AAMAS, AAAI, ICAPS, IJCAI) and reviewers at most major journals (JAIR, AIJ, JAAMAS) in Artficial Intelligence . He was nominated for best senior program committee member at AAMAS'13 and one of his papers was nominated for best student paper at AAMAS'09.
Host: Milind Tambe
Location: Ronald Tutor Hall of Engineering (RTH) - 526
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
Phd Defense - Jongwoo Lim
Thu, Sep 05, 2013 @ 12:00 PM - 01:30 PM
Thomas Lord Department of Computer Science
University Calendar
An Efficient Approach to Clustering Datasets with Mixed Type Attributes in Data Mining
PhD Candidate: Jongwoo Lim
Date and Time: 09/05/2013(Thr), 12:00pm ~ 1:30pm
Location: SAL 322
Prof. Dennis McLeod (Chairperson )
Prof. Aiichiro Nakano
Prof. Larry Pryor (Outside Member)
We propose an efficient approach to clustering datasets with mixed type attributes (both numerical and categorical), while minimizing information loss during clustering. Real world datasets such as medical datasets, bio datasets, transactional datasets and its ontology have mixed attribute type datasets.
However, most conventional clustering algorithms have been designed and applied to datasets containing single attribute type (either numerical or categorical). Recently, approaches to clustering for mixed attribute type datasets have emerged, but they are mainly based on transforming attributes to straightforwardly utilize conventional algorithms. The problem of such approaches is the possibility of distorted results due to the loss of information because significant portion of attribute values can be removed in the transforming process without knowledge background of datasets. This may result in a lower accuracy clustering.
To address this problem, we propose a clustering framework for mixed attribute type datasets without transforming attributes. We first utilize an entropy based measure of categorical attributes as our criterion function for similarity. Second, based on the results of entropy based similarity, we extract candidate cluster numbers and verify our weighting condition that is based on the degree of well balanced clusters with pre-clustering results and the ground truth ratio from the give dataset. Finally, we cluster the mixed attribute type datasets with the extracted candidate cluster numbers and the weights.
We have conducted experiments with a heart disease dataset and a German credit dataset, for which an entropy function as a similarity measure and the proposed method of extracting number of clusters has been utilized. We also experimentally explore the relative degree of balance of categorical vs. numerical attributes sub datasets in given datasets. Our experimental results demonstrate that the proposed framework improved accuracy effectively for the given mixed type attribute datasets.
Location: Henry Salvatori Computer Science Center (SAL) - 322
Audiences: Everyone Is Invited
Contact: Lizsl De Leon
-
USC ASCE First General Meeting
Thu, Sep 05, 2013 @ 05:00 PM - 06:30 PM
Viterbi School of Engineering Student Organizations
Student Activity
The American Society of Civil Engineers Student Chapter at the University of Southern California is a student-run organization that enriches the lives of civil and environmental engineering students through social events, community service, industry interaction, and engineering competitions. USC ASCE serves as the link between the university and ASCE & the professional experience.
USC ASCE was founded in 1924 and is the nation's third oldest student chapter. Our local chapter has bi-weekly meetings and social events that foster member interaction with other civil engineering students. Corporate guest speakers provide members with valuable professional advice, as well as opportunities to meet with experts from the civil engineering industry.
We also participate in the annual ASCE Pacific South West Conference (PSWC), where students from 18 universities compete in a variety of events, such as concrete canoe, steel bridge, and environmental design competitions. Our design teams spend much of the year preparing for PSWC, designing and building their projects. At the competition, we race the concrete canoes and the steel bridges are assembled and load-tested in a timed competition. Our environmental team took first place in their competition last year!
Please join us at our first general meeting to learn more about USC ASCE, fill out your membership form, and find out what we have planned for this year!
Check out our website for more information about USC ASCE: http://uscasce.com/
Like us on facebook: http://facebook.com/uscasce
Location: Kaprielian Hall (KAP) - 158
Audiences: Everyone Is Invited
Contact: American Society of Civil Engineers
-
BMEStart Informational Meeting
Thu, Sep 05, 2013 @ 08:00 PM - 09:00 PM
Viterbi School of Engineering Student Organizations
Student Activity
BMEstart is USC's sole medical device design team. Partnered with ASBME, BMEstart allows dedicated and motivated students to utilize their engineering and medical prowess to bring a biomedical device from inception to reality. Work with some of the brightest engineers on campus to tackle real world medical problems with intense brainstorming, hands-on learning, and rapid prototyping. Click HERE to get more information about what BMEStart is all about, and come to the meeting to see if you’d like to get involved.
Location: Mark Taper Hall Of Humanities (THH) - 119
Audiences: Everyone Is Invited
Contact: Associated Students of Biomedical Engineering
