User Tools

Site Tools


projects

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision
Previous revision
projects [2020/08/17 17:00] aanprojects [2020/08/17 17:27] (current) aan
Line 4: Line 4:
  
 Below is a list of 4080/4088 projects proposed by faculty members. More projects may be added once they come in (normally in August). Please contact the supervisor directly you are in interested in a project. You can also contact faculty members individually to discuss possible projects. Click [[http://eecs.lassonde.yorku.ca/community/faculty-members/|here]] to see a list of faculty members, their research areas and their contact information. Below is a list of 4080/4088 projects proposed by faculty members. More projects may be added once they come in (normally in August). Please contact the supervisor directly you are in interested in a project. You can also contact faculty members individually to discuss possible projects. Click [[http://eecs.lassonde.yorku.ca/community/faculty-members/|here]] to see a list of faculty members, their research areas and their contact information.
 +
 +====== Characterizing, Understanding, and Detecting Software Bugs in Machine Learning Frameworks ======
 +**Course:** EECS4080
 +
 +**Supervisor:** Song Wang 
 +
 +**Contact:** <wangsong@yorku.ca>
 +
 +Today, machine learning and its applications has been integrated into every part of our society. With the development and evolution of information technologies such as smartphones, self-driving cars, and robotics, the dependence of our society on machine learning systems continues to grow, which makes the impact and cost of bugs in these intelligent systems increase dramatically. Building reliable and secure machine learning systems has become an increasingly critical challenge for software developers. Thus, characterizing and understanding the root cause of software bugs in ML frameworks is critical to detecting and repairing buggy behavior.
 +
 +This project will first conduct an empirical study to analyze and summarize unique characteristics of software bugs collected from four typical ML frameworks (PyTorch, TensorFlow, CNTK, and Theano). Based on the findings from the empirical study, we will further propose actionable guidelines for designing and exploring new software bug detection approaches to find new bugs on machine learning frameworks and their applications.
 +
 +**Required skills:** General EECS4313 and EECS3311 prerequisites;
 +
 +**Recommended skills:** Solid programming skills with Python and Java.
  
 ====== Image Processing for Social Coding ====== ====== Image Processing for Social Coding ======
Line 23: Line 38:
  
 ====== Release Management for Mobile Apps ====== ====== Release Management for Mobile Apps ======
-**Course:** EECS4088/4480+**Course:** EECS4080
  
 **Supervisor:** Maleknaz Nayebi  **Supervisor:** Maleknaz Nayebi 
projects.1597683613.txt.gz · Last modified: 2020/08/17 17:00 by aan

Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki