Select a calendar:
Filter February Events by Event Type:
SUNMONTUEWEDTHUFRISAT
University Calendar
Events for February
-
PhD Thesis Proposal - Yuchen Lin
Tue, Feb 08, 2022 @ 02:00 PM - 03:30 PM
Thomas Lord Department of Computer Science
University Calendar
PhD Candidate: Yuchen Lin
Title: Commonsense Reasoning for Natural Language Processing
Tuesday: Feb 8, 02:00 PM - 03:30 PM
Committee members: chair: Prof. Xiang Ren (CS dept.), Prof. Cyrus Shahabi (CS dept.), Prof. Yan Liu (CS dept.), Prof. Robin Jia (CS dept.), Prof. Toby Mintz (Department of Psychology).
Abstract:
Common sense is all the background knowledge we have about the physical and social world that we have absorbed over our lives. It includes such things as our understanding of physics as well as our expectations about how humans behave. Commonsense knowledge is required before intelligent agents can anticipate how people and the physical world will react before they make decisions. However, it has long been a bottleneck for developing artificial general intelligence to use commonsense reasoning ability for understanding and generating natural language in real-world situations. In this thesis proposal, I present a series of work that enables natural language processing models to reason with commonsense knowledge. Specifically, my Ph.D. work focuses on using external knowledge bases to improve neural language models and developing generative, open-ended reasoning models that can serve real-life applications.
WebCast Link: https://usc.zoom.us/j/9502121213
Audiences: Everyone Is Invited
Contact: Lizsl De Leon
-
Ph.D. Thesis Proposal - Rajat Tandon
Thu, Feb 10, 2022 @ 12:00 PM - 01:00 PM
Thomas Lord Department of Computer Science
University Calendar
Ph.D. Thesis Proposal - Rajat Tandon
Thursday, February 10th, 2022: 12pm - 1pm
Title: Protecting online services from sophisticated attacks
Thesis Committee members: Dr. Jelena Mirkovic, Dr. Genevieve Bartlett, Dr. Emilio Ferrara, Dr. Ramesh Govindan, Dr. Chris Kyriakakis, Dr. Barath Raghavan
Abstract: Online services are often targets of sophisticated attacks, which aim to overwhelm services or steal user data. In this work, we present solutions, which aim to protect services against sophisticated distributed denial-of-service attacks. These solutions can effectively handle attacks that: (1) involve sending requests which resemble legitimate ones, (2) involve exploiting vulnerabilities that exist in different online services, (3) take advantage of the changing trends in network traffic, and (4) often require online services to get help from their ISPs for mitigation, due to the high volumes of attack traffic.
Zoom link: https://usc.zoom.us/j/98623683240WebCast Link: https://usc.zoom.us/j/98623683240
Audiences: Everyone Is Invited
Contact: Lizsl De Leon
-
Phd Thesis Proposal - Chaoyang He
Fri, Feb 11, 2022 @ 03:00 PM - 04:30 PM
Thomas Lord Department of Computer Science
University Calendar
Ph.D. Candidate: Chaoyang He
Title: Towards End-to-end Federated Machine Learning at Scale: Algorithm, Systems, and Applications
Committee members: Prof. Salman Avestimehr (Chair), Prof. Mahdi Soltanolkotabi, Prof. Murali Annavaram, Prof. Xiang Ren, Prof. Barath Raghavan
Abstract: Federated learning (FL) is a machine learning paradigm that many clients (e.g. mobile/IoT devices or whole organizations) collaboratively train a model under the orchestration of a central server (e.g. service provider), while keeping the training data decentralized. It has shown huge potential in mitigating many of the systemic privacy risks, regulatory restrictions, and communication costs, resulting from traditional, centralized machine learning and data science approaches in healthcare, finance, smart city, autonomous driving, and the Internet of things. Though it's promising, landing FL into trustworthy data-centric AI infrastructure faces many realistic challenges from learning algorithms (e.g., data heterogeneity, label deficiency) and distributed systems (resource constraints, system heterogeneity, security, privacy, etc.), requiring interdisciplinary research in machine learning, distributed systems, and security/privacy. Driven by this goal, My Ph.D. research focuses on end-to-end FL research, from algorithms to systems to applications. In this thesis proposal, I will first summarize my publications from the perspective of FedML, a widely adopted open-source library I developed, and highlight how I enable scale FL at cross-device and cross-silo settings, as well as diverse applications in CV, NLP, and data mining. Second, I will also introduce my broader accomplishment, including visionary papers, open-source impacts, academia service, industrial collaboration, invited talks, and workshop organization. Finally, I will briefly introduce my ongoing works on secure aggregation and label deficiency and finalize my presentation with a clear future plan.
Zoom Link: https://usc.zoom.us/my/usc.chaoyangheWebCast Link: https://usc.zoom.us/my/usc.chaoyanghe
Audiences: Everyone Is Invited
Contact: Lizsl De Leon
-
PhD Thesis Proposal - Liyu Chen
Fri, Feb 25, 2022 @ 02:00 PM - 03:00 PM
Thomas Lord Department of Computer Science
University Calendar
Time: 2:00-3:00pm Feb 25, Friday
Committee: Haipeng Luo (chair), Rahul Jain, David Kempe, Ashutosh Nayyar, Vatsal Sharan.
Title: Online Goal-Oriented Reinforcement Learning
Abstract: Reinforcement Learning (RL) studies how an agent learns to behave optimally in an unknown environment. It has been a popular topic in both industries and academia since AlphaGo demonstrated its great potential. However, there is still a large gap between theory and practice of RL due to the strong assumptions made in theoretical RL. My research focuses on online learning in a goal-oriented Markov Decision Process model named Stochastic Shortest Path (SSP), where the learner's objective is to reach a goal state with the smallest possible cost. Many real applications can be modeled by SSP such as games, car navigation, and robotic manipulations. To understand the SSP model better, we first focus on establishing minimax regret bounds in various settings. Specifically, for SSP with stochastic costs, we develop a simple minimax optimal algorithm concurrent to other works; for SSP with adversarial costs, we develop efficient minimax optimal algorithms with known transition, and near-optimal algorithms with unknown transition. Next, we focus on developing practical learning algorithms for SSP from different perspectives. Specifically, we develop the first model-free algorithm, the first set of policy optimization algorithms, and improved algorithms with linear function approximation.
For future work, I plan to study SSP for more general settings and develop more practical algorithms. For example, I plan to study the non-stationary SSP where both the transition and cost functions are changing, and SSP under general function approximation. I also plan to develop parameter-free SSP algorithms under different settings.WebCast Link: https://usc.zoom.us/j/97003272644
Audiences: Everyone Is Invited
Contact: Lizsl De Leon
-
PhD Defense - Shao-Hua Sun
Mon, Feb 28, 2022 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
University Calendar
PhD Candidate: Shao-Hua Sun
Time: 2/28 (Mon) 3:30pm PST
Venue: Online (zoom link: https://usc.zoom.us/j/2133002395)
Committee members: Prof. Joseph J. Lim (Chair), Prof. Gaurav Sukhatme, Prof. Stefanos Nikolaidis, Prof. Quan Nguyen (AME dept)
Title: Program-Guided Framework for Interpreting and Acquiring Complex Skills with Learning Robots
Abstract: Recent development in artificial intelligence and machine learning has remarkably advanced machines' ability to understand images and videos, comprehend natural languages and speech, and outperform human experts in complex games. However, building intelligent robots that can physically interact with their surroundings as well as learn to operate in unstructured environments, manipulate unknown objects, and acquire novel skills - to free humans from tedious or dangerous manual work- remains challenging. The focus of my research is to develop a robot learning framework that enables robots to acquire long-horizon and complex skills with hierarchical structures, such as furniture assembly and cooking. Specifically, I aim to devise a robot learning framework which is: (1) interpretable: by decoupling interpreting skill specifications (e.g. demonstrations, reward functions) and executing skills, (2) programmatic: by generalizing from simple instances to complex instances without additional learning, (3) hierarchical: by operating on a proper level of abstraction that enables human users to interpret high-level plans of robots allows for composing primitive skills to solve long-horizon tasks, and (4) modular: by being equipped with modules specialized in different functions (e.g. perception, action) which collaborate, allowing for better generalization. This dissertation discusses a series of research projects toward building such an interpretable, programmatic, hierarchical, and modular robot learning framework.
WebCast Link: https://usc.zoom.us/j/2133002395
Audiences: Everyone Is Invited
Contact: Lizsl De Leon
