Logo: University of Southern California

Events Calendar



Select a calendar:



Filter May Events by Event Type:



University Calendar
Events for May

  • USC AI Futures Symposium on Artificial Intelligence and Data Science

    Mon, May 03, 2021 @ 08:45 AM - 12:15 PM

    Information Sciences Institute, USC Viterbi School of Engineering

    University Calendar


    Profound innovations at the intersection of artificial intelligence and data science are changing our lives. These innovations are transforming how we improve our health, connect with others, sustain our environment, understand complex systems, and enrich our lives. This symposium will present an overview of interdisciplinary research at USC in these critical areas.

    This event is part of the USC AI Futures Symposium Series. A prior event was held in January 2021 with the theme: Will AIs Ever Be One of Us?.

    The sessions will run from 8:45am PST to 12:15 PST on May 3-5, 2021.

    Find out more: https://isi-usc-edu.github.io/USC-AI-DS-Symposium/

    Registration: https://usc.zoom.us/webinar/register/WN_V-mMUlHGQMWnkecKyPUQWA

    Audiences: Everyone Is Invited

    Contact: Yolanda Gil

    OutlookiCal
  • PhD Defense - Chung Ming Cheung

    Fri, May 07, 2021 @ 09:00 AM - 11:00 PM

    Computer Science

    University Calendar


    PhD Candidate: Chung Ming Cheung

    Time: May 7th 2021, 9 - 11 am

    Zoom link:
    prasannaseminars.github.io

    Committee: Professor Viktor K Prasanna (Chair)
    Professor C S Raghavendra
    Professor Aiichiro Nakano

    Title: Data-Driven Methods for Increasing Real-Time Observability in Smart Distribution Grids

    Abstract:
    Traditional power distribution grids have evolved into smart grids with the development of advanced metering infrastructures and renewable energy based distributed energy resources (DER). This has introduced the following challenges: (1) The stochasity of renewable energy based DERs has increased the volatility of grid frequency; (2) the decentralization of generation into small scaled DERs has reduced grid inertia. To address these challenges, real-time knowledge and understanding of signal measurements of grid assets, called observability, are crucial to make grid operation decisions swiftly. High observability can be obtained through extensive metering of assets in smart grids for data collection, and time series analytics that extract information from the collected time series data. However, the proliferation of DERs has introduced new challenges in these analytics. DERs located behind-the-meters (BTM) are not recorded individually and hidden from real-time observations. This combined with the volatile nature of DER assets greatly reduces observability. As a result, these data-driven models do not have full observability of data and suffer from accuracy losses.

    In this thesis, we develop data-driven approaches to improve observability. We develop unsupervised disaggregation models for separation of signals of BTM DERs hidden from net meter measurements. We focus on the separation of signals from the activity of BTM solar photovoltaics and battery storages. We also propose capturing spatial features using machine learning models such as spatial-temporal graph convolution networks for improving time series analytics in smart grids, e.g. load forecasting and missing data imputation. Moreover, we show that the increase in observability provided by these data-driven models can enhance other time series analytics in smart grids.

    WebCast Link: prasannaseminars.github.io

    Audiences: Everyone Is Invited

    Contact: Lizsl De Leon

    OutlookiCal
  • PhD Defense - Nazanin Alipourfard

    Fri, May 07, 2021 @ 09:00 AM - 11:00 AM

    Computer Science

    University Calendar


    PhD Candidate: Nazanin Alipourfard

    Date: May 7, 2021

    Time: 9-11am

    Dissertation defense committee:
    Kristina Lerman (chair), Ellis Horowitz, Jose-Luis Ambite, Greg Ver Steeg, Phebe Vayanos

    Title:
    Emergence and Mitigation of Bias in Data and Networks

    Abstract:
    The presence of bias often complicates the quantitative analysis of large-scale heterogeneous or network data. Discovering and mitigating these biases enables a more robust and generalizable analysis of data. This thesis focuses on the 1) discovery, 2) measurement and 3) mitigation of biases in heterogeneous and network data.

    The first part of the thesis focuses on removing biases created by the existence of diverse classes of individuals in the population. I describe a data-driven discovery method that leverages Simpson's paradox to identify subgroups within a population whose behavior deviates significantly from the rest of the population. Next, to address the challenges of multi-dimensional heterogeneous data analysis, I propose a method that discovers latent confounders by simultaneously partitioning the data into fuzzy clusters (disaggregation) and modeling the behavior within them (regression).

    The second part of this thesis is about biases in bi-populated networked data. First, I study the perception bias of individuals about the prevalence of a topic among their friends in the Twitter social network. Second, I show the existence of power-inequality in author citation networks in six different fields of study, due to which authors from one group (e.g., women) receive systematically less recognition for their work than another group (e.g., men). As the last step, I connect these two concepts (perception bias and power-inequality) in bi-populated networks and show that while these two measures are highly correlated, there are some scenarios where there is a disparity between them.

    Zoom Link:
    https://usc.zoom.us/j/93756467657?pwd=dWxEMHVMYnppZnAyZHRYVEVaTkZSQT09

    WebCast Link: https://usc.zoom.us/j/93756467657?pwd=dWxEMHVMYnppZnAyZHRYVEVaTkZSQT09

    Audiences: Everyone Is Invited

    Contact: Lizsl De Leon

    OutlookiCal
  • PhD Defense - Nitin Kamra

    Mon, May 10, 2021 @ 10:30 AM - 12:00 PM

    Computer Science

    University Calendar


    PhD Candidate: Nitin Kamra

    Committee: Prof. Yan Liu (chair), Prof. Bistra Dilkina, Prof. Ashutosh Nayyar

    Date: 10th May, 2021
    Time: 10:30am-12:00pm

    Zoom: https://usc.zoom.us/j/96990524233?pwd=M3BzcTdOenZtRjlkN1J5dmxDQmVpUT09

    Meeting ID: 969 9052 4233
    Passcode: 015551


    Title: Machine Learning in Interacting Multi-agent Systems

    Abstract:

    Making predictions and learning optimal behavioral strategies are important problems in many domains such as traffic prediction, pedestrian tracking, financial investments and security systems. These systems often consist of multiple agents interacting with each other in complex ways, which makes both the above tasks very challenging. In this thesis, I propose methods to advance the state-of-the-art for such multi-agent learning problems. The first part of my talk focuses on trajectory prediction and I will present a relational model involving a fuzzy decision making attention mechanism for multi-agent trajectory prediction. Our approach shows significant performance gains over many existing state-of-the-art predictive models in diverse domains such as human crowds, US freeway traffic and various physics datasets. The second part of my talk focuses on placing multiple resources to protect and cover geographical spaces. We propose the Coverage Gradient Theorem and a spatial discretization based framework to improve existing benchmarks for spatial coverage domains. The third part of the talk focuses on computing nash equilibrium strategies in spatial security games with continuous action spaces. We present our model-free learning algorithm, OptGradFP, and our model-based learning algorithm, DeepFP, which search for the optimal defender strategy in a parameterized continuous search space. These algorithms scale to large domains and compute strategies robust to adversarial exploitation. Finally, we combine the Coverage Gradient framework with DeepFP to show improved performance on spatial coverage security domains.

    WebCast Link: https://usc.zoom.us/j/96990524233?pwd=M3BzcTdOenZtRjlkN1J5dmxDQmVpUT09

    Audiences: Everyone Is Invited

    Contact: Lizsl De Leon

    OutlookiCal
  • Thesis Proposal - Shen Yan

    Mon, May 10, 2021 @ 11:00 AM - 01:00 PM

    Computer Science

    University Calendar


    Title: Fair Machine Learning for Human Behavior Understanding

    Time: 11:00 AM-1:00 PM PST, May 10 (Monday)

    Committee: Emilio Ferrara, Cyrus Shahabi, Shri Narayanan, Kristina Lerman, and Fred Morstatter.

    Zoom link: https://usc.zoom.us/j/96050343860

    Abstract:
    Artificial intelligence (AI) and machine learning models have been recently applied extensively to understand and predict human behavior, often in applications with major societal implications, such as making recruitment decisions, estimating daily well-beings, or assessing clinical treatments. Despite the increasing body of research on modeling human behavior and fair machine learning, most studies focus on homogeneous and objective measurements, and little has been discussed on how to mitigate the impact of heterogeneity on utility and fairness simultaneously. The increasing amount of collected data also raises concerns on data privacy. Recent regulations such as European General Data Protection Regulation (GDPR) and California Consumer Privacy Act (CCPA) regulate the usage of personal data. However, most previous fairness work requires the access of sensitive attributes (e.g., race, gender) to debias the system.
    This dissertation proposal will articulate the challenges posed by complex, multimodal human behavior data in both model utility and fairness. The proposed work is decomposed into three tasks, namely tackling machine learning fairness issues originating from the heterogeneous human behaviors (Task 1), and biased behavior annotations (Task 2), and designing fair machine learning methods without sensitive attributes (Task 3) for both centralized and federated learning. This work will provide possible solutions to mitigate bias of human behavior understanding systems, reducing barriers to access, alleviating systemic racism, discrimination, and unfair process.

    WebCast Link: https://usc.zoom.us/j/96050343860

    Audiences: Everyone Is Invited

    Contact: Lizsl De Leon

    OutlookiCal
  • Thesis Proposal - Jiaoyang Li

    Mon, May 10, 2021 @ 02:00 PM - 03:30 PM

    Computer Science

    University Calendar


    Title:

    Efficient and Effective Multi-Agent Path Finding via Heuristic Search, Symmetry Breaking, and Large Neighborhood Search


    Time:

    2:00-3:30pm PST, May 10 (Monday)

    Committee: Sven Koenig, Nora Ayanian, Bistra Dilkina, T. K. Satish Kumar, Satyandra K. Gupta, and Brian Williams (MIT)

    Zoom link:

    https://usc.zoom.us/j/96942890548?pwd=SkFzQm80cHNOMHBPQXZBa1N3eVZvZz09


    Abstract:

    Recent advances in robotics have laid the foundation for building large-scale multi-agent systems. One fundamental task in many multi-agent systems is to navigate teams of agents in a shared environment to their target locations without colliding with obstacles or other agents. Applications include evacuation, formation control, object transportation, traffic management, search and rescue, autonomous driving, drone swarm coordination, video game character control, and large-scale warehouse automation, to list a few. One well-studied abstract model for this problem is known as Multi-Agent Path Finding (MAPF). MAPF is NP-hard to solve optimally (or, in some cases, even bounded-suboptimally) in general. Existing algorithms for solving MAPF either have limited scalability, generate costly solutions, or fail to find any solutions for hard MAPF problems. I propose to develop fundamental techniques for solving MAPF more efficiently and effectively that exploit the combinatorial structure of the MAPF problem and combine ideas from both AI and OR. In particular, I design admissible heuristics by cost partitioning (ideas from the AI planning community) to speed up optimal MAPF algorithms. I also develop symmetry-breaking constraints (ideas from the constraint programming community) to eliminate symmetries in optimal MAPF solving. For future work, I plan to learn inadmissible heuristics (ideas from the heuristic search community) to speed up bounded-suboptimal MAPF algorithms and build an anytime MAPF framework via large neighborhood search (ideas from the constraint programming and operations research communities) to improve the success rate and the solution quality of non-optimal MAPF algorithms.

    WebCast Link: https://usc.zoom.us/j/96942890548?pwd=SkFzQm80cHNOMHBPQXZBa1N3eVZvZz09

    Audiences: Everyone Is Invited

    Contact: Lizsl De Leon

    OutlookiCal
  • PhD Defense - Ryan Julian

    Tue, May 11, 2021 @ 04:00 PM - 05:30 PM

    Computer Science

    University Calendar



    PhD Candidate: Ryan Julian

    Committee: Prof. Gaurav Sukhatme (chair), Prof. Satyandra Gupta, Prof. Heather Culbertson, Prof. Joseph Lim, Prof. Stefanos Nikolaidis, Dr. Karol Hausman

    Title: Algorithms and Systems for Continual Robot Learning

    Abstract:
    The last decade has seen the rapid evolution of machine learning (ML) from an academic curiosity to an essential capability of modern computing systems, and with it has come an explosion of ML applications which help humans in the real world every day. Whether machine learning can affect such an evolution in the nearby field of robotics--concerned with automating tasks in the physical, rather than the digital world--remains to be seen. Though the last decade has seen an explosion of new methods, remarkably few of these exhibit properties which make them good candidates for building an efficient, continual, multi-task robot skill learning system. This of course begs the question: given a novel method, how should we assess its suitability for building a continually-learning multi-task robot? In other words, how can we systematically assure progress towards new real world capabilities, and avoid getting trapped, running in circles studying novel trivialities? I will argue that three elements--benchmarks, baselines, and novel methods--together form the three-legged stool of research in artificial intelligence. Robotics can only make progress towards the goal of real-world, general-purpose, continually-learning robots by periodically advancing each of these legs, creating a virtuous cycle of new challenges, new systems, new solutions, and ultimately new knowledge. We will follow one such cycle for a small slice of the field, namely continual learning for robotic manipulation, visiting each step to illustrate its effectiveness by example.

    Zoom Info:
    Topic: PhD Defense - Ryan Julian
    Time: May 11, 2021 04:00 PM Pacific Time (US and Canada)

    Join Zoom Meeting
    https://usc.zoom.us/j/94821577914?pwd=NW14YW5WWjdPcXVna3grcjZqQnFuQT09

    WebCast Link: https://usc.zoom.us/j/94821577914?pwd=NW14YW5WWjdPcXVna3grcjZqQnFuQT09

    Audiences: Everyone Is Invited

    Contact: Lizsl De Leon

    OutlookiCal
  • PhD Defense - Sivaram Ramanathan

    Thu, May 13, 2021 @ 11:00 AM - 01:00 PM

    Computer Science

    University Calendar


    Ph.D. Candidate: Sivaram Ramanathan

    Committee: Dr. Jelena Mirkovic, Dr. Minlan Yu, Dr. Emilio Ferrara and Dr. Ramesh Govindan

    Time: May 13, 2021, 11am

    Title: Improving Network Security and Performance Through Programmability and Machine Learning

    Abstract:
    The rise in different types of applications has attracted many users to the Internet. Companies generate revenue from users and a key component for user retention is reliable network performance. Network operators are constantly scaling their infrastructure to provide tight network and security guarantees to their users. However, issues in networks such as packet drops, low utilization of links, and targeted attacks can violate these guarantees.

    Network operators use different tools to understand and diagnose problems in the network. As the network scales to support more users, tools that are traditionally used to understand and diagnose problems, also need to change. For instance, there exist transient events occurring at microsecond granularity in datacenter networks that could affect the network's performance. Traditional tools may miss such events as they work at coarser time granularities.

    As networks grow to accommodate more users, securing the network has also become hard. In the past year, there has been a 776% increase in large volumetric denial of service (DDoS) attacks and networks have spent up to $50,000 to protect themselves. Moreover, most deployed defenses are reactive, where a mitigation strategy is only developed when symptoms of attacks are seen. Proactively detecting attackers would not only block all attack traffic but also reduce cost for victim networks.

    In this talk, we use recent advancements in programmable switches and machine learning to develop frameworks for better network management. We present SPred, which uses machine learning models in switches to detect transient events faster. We designed SDProber to balance the cost of monitoring with event detection time. We also built frameworks that help network operators to meet security guarantees. We present SENSS, which allows networks to coordinate with upstream networks to develop better detection and mitigation strategies against DDoS attacks. Finally, we present BLAG that makes blocklists more suitable for emergency response by combining blocklists of different attack types and reducing the collateral damage by using a recommendation system and reused address detection.

    Our work has had several real-world impacts. SENSS has been deployed in three academic networks to provide better detection of DDoS attacks. Our technique to identify reused addresses is being adopted by IPInfo, which maintains a large repository of IP address-related information. Finally, AT&T has partially deployed SDProber in their network to detect persistent congestion and we hold two patents for SDProber and SPred.

    Zoom info:
    https://usc.zoom.us/j/98493286938?pwd=UXRFVzNlQUl5aVJtUXNQZnpEekNaZz09


    Topic: Sivaram's Defense

    Join Zoom Meeting
    https://usc.zoom.us/j/98493286938?pwd=UXRFVzNlQUl5aVJtUXNQZnpEekNaZz09

    WebCast Link: https://usc.zoom.us/j/98493286938?pwd=UXRFVzNlQUl5aVJtUXNQZnpEekNaZz09

    Audiences: Everyone Is Invited

    Contact: Lizsl De Leon

    OutlookiCal
  • CEE 2021 Virtual Commencement Celebration

    Thu, May 13, 2021 @ 02:30 PM - 03:00 PM

    Sonny Astani Department of Civil and Environmental Engineering

    University Calendar


    To honor and celebrate the Class 2020-21 Civil and Environmental Engineering graduates via zoom, please click on the following link:

    https://usc.zoom.us/j/97944495184?pwd=ZXN5d0FPdk1XcEdBdnl6dHZYaHhjQT09
    Passcode: 332144

    Location: ONLINE

    WebCast Link: https://usc.zoom.us/j/97944495184?pwd=ZXN5d0FPdk1XcEdBdnl6dHZYaHhjQT09

    Audiences: Everyone Is Invited

    Contact: Salina Palacios

    OutlookiCal
  • PhD Defense - Mian Wan

    Fri, May 14, 2021 @ 01:30 PM - 04:00 PM

    Computer Science

    University Calendar


    PhD Candidate: Mian Wan



    Date: May 14th, 2021

    Time: 1:30-4pm

    title: Automatic Detection and Optimization of Energy Optimizable UIs in Android Applications Using Program Analysis


    Committee: Prof. William Halfond (chair), Prof. Nenad Medvidovic, Prof. Chao Wang, Prof. Jyotirmoy Deshmukh, Prof. Sandeep Gupta

    Zoom link:
    https://usc.zoom.us/j/94745439598

    Abstract:
    Mobile apps and smartphones play an essential role in our daily life, and the energy consumption of an app has become an important concern for its developers. Given the fact that an app's display energy consumption can be optimized at the software level, many techniques have been proposed to help optimize the apps' display energy on OLED screens. However, there are no automated techniques for detecting and repairing energy optimizable user interfaces (UIs) in Android apps. Instead, for detection, the developers can only manually examine each UI's colors and determine which UIs are optimizable based on their intuition. As for repairing, the developers need to manually analyze the app to modify the color settings to recolor the UIs.

    My dissertation overcomes the above challenges and limitations by automating the process of detecting and repairing energy optimizable UIs in mobile apps. I have two main insights underlying my research. The first insight is that display energy optimization potential can be quantified. Based on this insight, I designed and developed an approach that combines dynamic analysis, power modeling, and color transformation to detect energy optimizable UIs. The second insight is that both types of program analyses can be used to gather UI information. Based on this insight, I designed and developed an approach that employs dynamic analysis, static analysis, and a search based technique to model and recolor Android UIs. In the empirical evaluation, my techniques were highly effective and efficient in detecting and repairing the energy optimizable UIs in mobile apps. These results indicate that my detection technique can help developers in locating energy optimizable UIs and that my repair technique can help developers in repairing energy optimizable UIs while maintaining their aesthetic quality.

    WebCast Link: https://usc.zoom.us/j/94745439598

    Audiences: Everyone Is Invited

    Contact: Lizsl De Leon

    OutlookiCal
  • Global Accessibility Awareness Day (GAAD)

    Thu, May 20, 2021

    Information Technology Program (ITP)

    University Calendar


    Thursday, May 20th is the tenth annual Global Accessibility Awareness Day (GAAD). The purpose of GAAD is to get everyone talking, thinking and learning about digital access and inclusion, and the more than One Billion people with disabilities/impairments. ITP advisory Board Member Joe Devon, one of the co-founders of GAAD, created this video for the Trojan community: https://drive.google.com/file/d/1mVBY4iBB0qA4pAhlCOSeO1ZCS1t0MJtj/view?usp=sharing sharing the story of how GAAD began.

    ITP Faculty Member Kendra Walther is actively involved in Viterbi's accessibility efforts and serves as co-lead of the Teach Access (https://teachaccess.org/) student task force and co-organizer of the 2021 Teach Access Virtual Study Away Program. During the program, over 80 students from nine universities participated in sessions with accessibility experts from a variety of technology companies, and 13 of those student participants were Trojans. Students learned about accessibility and assistive technologies, heard from disability advocates, learned about career paths in accessibility, engaged in AI and VR research in accessibility, discussed racial justice, intersectionality, and disability rights, thought about accessible design and hosting accessible events, and had the opportunity to apply their knowledge in an optional team project. Participating students were placed in small cross-disciplinary, multi-institutional teams and competed to build their own accessibility focused awareness or technology project during the Accessathon. Winners will be announced on the Teach Access website on GAAD, and each team has at least one Trojan student member.

    On this GAAD, we encourage everyone in the Trojan family to talk, think, and learn about accessibility and inclusion! Fight on!

    Audiences: Everyone Is Invited

    Contact: Eric Perez

    OutlookiCal