Sina Ghiassian

Department of Computing Science · University of Alberta · Edmonton, Alberta · ghiassia@ualberta.ca

My goal is to understand computational principles underlying intelligence. I am interested in the broad area of Artificial Intelligence, specifically in Reinforcement Learning and how machines learn to have a goal-directed behavior by interacting with the world. My primary research focus is on understanding off-policy learning algorithms and building new stable algorithms that learn fast. A second topic of interest is fully incremental deep reinforcement learning through mitigating catastrophic interference.


PUBLICATIONS

REFEREED JOURNAL PUBLICATIONS

Using Functional or Structural Magnetic Resonance Images and Personal Characteristic Data to Identify ADHD and Autism

Sina Ghiassian, Russell Greiner, Ping Jin, Matthew R.G. Brown. PLOS One1234 11.12 (2016): e0166934.
2016

REFEREED CONFERENCE PAPERS

Gradient Temporal-Difference Learning with Regularized Corrections.

Sina Ghiassian, Andrew Patterson, Shivam Garg, Dhawal Gupta, Adam White, Martha White. Accepted to the International Conference on Machine Learning 2020 (ICML 2020).
2020

Improving Performance in Reinforcement Learning by Breaking Generalization in Neural Networks

Sina Ghiassian, Banafsheh Rafiee, Yat Long Lo, Adam White. In Proceedings of the 19th International Conference on Autonomous Agents and MultiAgent Systems (AAMAS~2020), pp. 438-446.
2020

Prediction in Intelligence: An Empirical Comparison of Off-policy Algorithms on Robots.

Banafsheh Rafiee, Sina Ghiassian, Adam White, Richard Sutton. In Proceedings of the 18th International Conference on Autonomous Agents and MultiAgent Systems (AAMAS 2019), pp. 332-340.
2020

REFEREED WORKSHOP PAPERS

Overcoming Catastrophic Interference in Online Reinforcement Learning with Dynamic Self-Organizing Maps

Yat Long Lo, Sina Ghiassian. Accepted to the Biological and Artificial Reinforcement Learning workshop of the Neural Information Processing Systems conference in Nov. 2019.
2019

Robot Off-policy Prediction: An empirical Comparison of Learning Algorithms.

Banafsheh Rafiee, Sina Ghiassian, Adam White, Richard Sutton. Accepted to the Continual Learning workshop of the Neural Information Processing Systems conference (NIPS CL) in Nov. 2018.
2018

Rejection Sampling for Off-Policy Reinforcement Learning.

Wesly Chung, Sina Ghiassian, Somjit Nath, Martha White. Accepted to the Continual Learning workshop of the Neural Information Processing Systems conference (NIPS CL) in Nov. 2018.
2018

A First Empirical Study of Emphatic Temporal Difference Learning.

Sina Ghiassian, Banafsheh Rafiee, Richard Sutton. Accepted to the Continual Learning workshop of the Neural Information Processing Systems conference (NIPS CL) in Nov. 2016.
2016

Learning to Clasify Psychiatric Disorders based on fMR Images: Autism vs. Healthy and ADHD vs. Healthy.

Sina Ghiassian, Russell Greiner, Ping Jin, Matthew R.G. Brown. Accepted to the Machine Learning and Interpretation in NeuroImaging workshop of the Neural Information Processing Systems conference (NIPS MLINI) in Nov. 2013.
2013

Education

University of Alberta, Edmonton, Alberta, Canada

Ph.D. Candidate, Computer Science, GPA: 4.3/4
Specialization: Reinforcement Learning
Supervisor: Richard S. Sutton, Adam White
Sep. 2015 – present

University of British Columbia, Vancouver, British Columbia, Canada

Ph.D. Student, Neuroscience (voluntary withdrawal)
Specialization: fMRI Data Analysis
Supervisor: Todd S. Woodward
Sep. 2014 – Aug. 2015

University of Alberta, Edmonton, Alberta, Canada

M.Sc. Computer Science, GPA: 4.1/4
Specialization: Supervised Learning
Supervisor: Russell Greiner, Matthew R.G. Brown
Sep. 2012 – Aug. 2014

TEACHING ASSISTANT

Reinforcement learning for Artificial Intelligence.

Department of Computing Science, University of Alberta.
Sep. 2017 – Dec. 2017

Advanced Python Programming.

Department of Computing Science, University of Alberta.
Jan. 2013 – Apr. 2013

Introductory Python Programming.

Department of Computing Science, University of Alberta.
Sep. 2012 – Dec. 2012

HONOURS AND AWARDS

  • J.P. Morgan (AI) Ph.D. Fellowship (international fellowship) - Value: $55000 – at University of Alberta
  • Queen Elizabeth 2 Scholarship (provincial scholarship) - Value: $15000 – at University of Alberta.
  • International Tuition Award - Value: $1066 – at University of British Columbia.
  • Faculty of Medicine Graduate Award - Value: $3500 – at University of British Columbia.
  • International Tuition Award - Value: $2133 – University of British Columbia.
  • Graduate Student GPA Award - Value: $1000 – University of Alberta.