Select a calendar:
Filter June Events by Event Type:
SUNMONTUEWEDTHUFRISAT
University Calendar
Events for June
-
PhD Defense - Jiaping Gui
Fri, Jun 15, 2018 @ 09:00 AM - 11:30 AM
Thomas Lord Department of Computer Science
University Calendar
Title: Utilizing User Feedback to Assist Software Developers to Better Use Mobile Ads in Apps
PhD Candidate: Jiaping Gui
June 15th
9am - 11:30am
SAL 322
Committee:
William G.J. Halfond (Chair)
Nenad Medvidovic
Chao Wang
Jyotirmoy Deshmukh
Paul Bogdan (EE department, outside member)
Meiyappan Nagappan (U of Waterloo, outside member)
Abstract:
In the mobile app ecosystem, developers receive ad revenue by placing ads in their apps and releasing them for free. While there is evidence that users do not like ads, we do not know what are the aspects of ads that users dislike nor if they dislike certain aspects of ads more than the others. Therefore, in the first piece of my dissertation work, I analyzed the different ad related topics of ad reviews from users. In order to do this, I investigated app reviews that users gave for apps in the app store that were about ads. I found that most ad complaints were about UI related topics and three topics discussed predominantly were: the frequency with which ads were displayed, the timing of when ads were displayed, and the location of the displayed ads. I also found users reviewed non UI aspects of mobile advertising, such as ads blocking or slowing down the host app's running. Then I quantified different ad metrics corresponding to both UI and non UI ad aspects that were complained about most by end users. In the end, I correlated these quantified ad aspects with app ratings. For non UI ad aspects, I found that complaints about these aspects were significant and could impact the ratings (on a scale of one to five stars) given to an app. For UI ad aspects, I found that lower ratings (with statistical significance) were generally associated with apps that had different visual patterns regarding ad implementation, such as ads in the middle or bottom of the page, and ads in the initial landing page of an app, Based on the results, a set of guidelines were distilled to help app developers more effectively use ads in their apps.
Location: Henry Salvatori Computer Science Center (SAL) - 322
Audiences: Everyone Is Invited
Contact: Lizsl De Leon
-
PhD Defense - Yazeed Alabdulkarim
Thu, Jun 21, 2018 @ 09:00 AM - 11:00 AM
Thomas Lord Department of Computer Science
University Calendar
Title: Polygraph: A Plug-n-Play Framework to Quantify Application Anomalies
Date and Time: June 21, 9-11am in SAL 213
Phd Candidate: Yazeed Alabdulkarim
Committee:
Professor Shahram Ghandeharizadeh (chair)
Professor Chao Wang
Professor Christopher Gould (outside member)
Title: Polygraph: A Plug-n-Play Framework to Quantify Application Anomalies
Date and Time: June 21, 9-11am in SAL 213
Abstract:
Polygraph is a plug-n-play tool to quantify application anomalies. Example anomalies include erroneous results attributed to a software bug or an incorrect implementation, dirty reads when a database management system is configured with a weak consistency setting, and stale data produced by a cache. Polygraph is application agnostic and scales to detect anomalies and compute freshness confidence in real-time. It consists of an authoring, monitoring, and validation components. An experimentalist uses Polygraph authoring tool to generate code snippets to embed in the application software. These code snippets generate log records capturing conceptual application transactions at the granularity of entities and their relationships. Polygraph validates application transactions in a scalable manner by dividing the task into sub-tasks that process fragments of log records concurrently. Polygraph's monitoring tool visualizes transactions to help reason about anomalies. Polygraph is extensible, enabling a developer to tailor one or more of its components to be application specific. For example, we extend Polygraph's validation component to validate range predicates and simple analytics, such as max and min. We demonstrate the use of Polygraph with a variety of benchmarks representing diverse applications including TPC-C, BG, YCSB, SEATS and TATP.
Location: Henry Salvatori Computer Science Center (SAL) - 213
Audiences: Everyone Is Invited
Contact: Lizsl De Leon
-
PhD Defense - Wei-Lun Chao
Tue, Jun 26, 2018 @ 04:00 PM - 06:00 PM
Thomas Lord Department of Computer Science
University Calendar
PhD Candidate: Wei-Lun Chao
Title: Transfer Learning for Intelligent Systems in the Wild
Committee:
Professor Fei Sha (chair)
Professor Laurent Itti
Professor Joseph Lim
Professor Jason Lee (outside member)
Professor Panayiotis Georgiou (outside member)
Date and Time: June 26, 4-6 pm in SAL 322
Abstract:
Developing intelligent systems for vision and language understanding has long been a crucial part that people dream about the future. In the past few years, with the accessibility to large-scale data and the advance of machine learning algorithms, vision and language understanding has had significant progress for constrained environments. However, it remains challenging for unconstrained environments in the wild where the intelligent system needs to tackle unseen objects and unfamiliar language usage that it has not been trained on. Transfer learning, which aims to transfer and adapt the learned knowledge from the training environment to a different but related test environment has thus emerged as a promising paradigm to remedy the difficulty.
My thesis focuses on two challenging paradigms of transfer learning: zero-shot learning and domain adaptation. I will begin with zero-shot learning, which aims to expand the learned knowledge from seen objects, of which we have training data, to unseen objects, of which we have no training data. I will present an algorithm SynC that can construct the classifier of any object class given its semantic description, even without training data, followed by a comprehensive study on how to apply it to different environments. I will then describe an adaptive visual question answering framework that builds upon the insight of zero-shot learning and can further adapt its knowledge to new environments given limited information.
Location: Henry Salvatori Computer Science Center (SAL) - 322
Audiences: Everyone Is Invited
Contact: Lizsl De Leon
