The Scientist

scientist

I successfully defended my PhD at Northeastern University , under the advisement of Javed Aslam.

I am a researcher with expertise in multi-armed bandits and general machine learning. I am particularly interested in working in fields like recommender systems, personalization and advertisement. I also have an interest in speeding up training algorithms under budget (or time) constraints. For my PhD thesis, I developed bandit model algorithms for exploring and choosing good options when you have unlimited possibilities. I was the first to prove upper and lower bounds on the budget needed to find one of the top few of unlimited options, and my fixed-budget algorithm takes a surprising approach that empirically matches or outperforms all state-of-the-art algorithms. I have worked on applications like training greedy optimal boosted decision trees faster and finding the optimal dose in cancer clinical trials. I am also interested in pivoting to nearby fields, such as reinforcement learning.

Current project: I am currently wrapping up some work in the pure exploration setting of the infinitely-armed bandits in the fixed budget setting. This work is in collaboration with Kevin Jamieson and Javed Aslam.

Preprints/Under Review:

  • On Multi-Armed Bandit Designs for Phase I Clinical Trials.
    Maryam Aziz, Emilie Kaufmann, Marie-Karelle Riviere. PDF

  • Adaptively Pruning Features for Boosted Decision Trees.
    Maryam Aziz, Jesse Anderton, and Javed Aslam. PDF

Publications:

  • Pure Exploration in Infinitely-Armed Bandit Models with Fixed-Confidence.
    Maryam Aziz, Jesse Anderton, Emilie Kaufmann and Javed Aslam.  ALT 2018. PDF

  • Health Literacy and Usability of Clinical Trial Search Engines.
    Timothy Bickmore, Michael Paasche-Orlow, Maryam Aziz, Barbara Barry. Proc. Health Literacy Annual Research Conference (HARC) 2012.

  • Using Computer Agents to Improve Emotional Wellbeing of Hospital Patients.
    Maryam Aziz, Timothy Bickmore, Laura Vardoulakis, Christopher Shanahan, Michael Paasche-Orlow. CHI workshop on Interaction Design and Emotional Wellbeing 2012. PDF

Research Experience:

  • Fall 2012 – Present. Machine Learning Lab, CCIS, Northeastern University. Advised by Javed Aslam.
  • Summer of 2016. Inria Lille, SequeL team. Worked with Emilie Kaufmann.
  • Fall 2011 – Fall 2012. Relational Agents Lab, CCIS, Northeastern University. Mentored by Timothy Bickmore.

Teaching at Northeastern University:

  • CS1100: Computer Science and Its Applications:  Fall 2014
  • CS1100: Computer Science and Its Applications: Fall 2012

Teaching Assistantship at Northeastern University:

  • CS6200: Information Retrieval: Spring 2017
  • CS6200: Information Retrieval: Fall 2016
  • CS6120: Natural Language Processing: Spring 2016
  • CS2510: Introduction to Class-based Program Design: Fall 2015
  • CS6200: Information Retrieval: Spring 2015
  • CS6200: Information Retrieval: Summer 2014
  • CS2510: Introduction to Class-based Program Design: Spring 2014
  • CS4100/5100: Foundations of Artificial Intelligence: Fall 2013
  • CS3200: Database Design: Spring 2013