Alina Vereshchaka

PhD Student, Research Assistant in Artificial Intelligence

Fall 2019 Office Hours: Tue/Thu 2:00 - 3:00pm

Click here for CSE4/510 RL course page (Fall 2019)

avereshc[at]buffalo.edu

About

I am a PhD candidate advised by Professor Wen Dong in the Computer Science and Engineering Department at the University at Buffalo. I have been deeply working in Reinforcement Learning, and my current research focuses on optimal control in complex systems.

Education

Fall 2017 - Present
PhD student in Computer Science and Engineering
University at Buffalo, the State University of New York (SUNY), NY, USA

Academic Teaching Experience

Instructor of Record, University at Buffalo -- Fall 2019
Instructor of Record, University at Buffalo -- Summer 2019
Teaching Assistant, University at Buffalo

Publications

Tutorials

Awards

Professional Service

Projects

Predicting human behaviors from electroencephalography

Tools & Algorithms: Python3, VAE, NN, Matplotlib
Note: The proposed approach and preliminary results won the second place award at IEEE Brain Data Bank Challenge & Competition, Seattle, USA, Dec 2018

Stabilizing policy optimization by constraining gradient updates in deep reinforcement learning

Tools & Algorithms: Python3, Keras, Tensorflow, Stable Baseline, OpenAI Gym, PPO, TRPO

Learning to navigate in complex environment using deep reinforcement learning

Tools & Algorithms: OpenAI Gym, Keras-RL, CNN, DQN
Note: adapted version of the project was used a main course project in Machine Leaning course at University at Buffalo

Deep residual learning for NIST images recognition (code)

Tools & Algorithms: Keras, CNN

Digits classification (MNIST) using various machine learning and deep learning algorithms

Tools & Algorithms: Python, Numpy, Tensorflow, Numpy, Matplotlib, SVM, CNN, NN

Probabilistic Graphical Models (code)
Making exact inferences about probabilistic graphical models, by constructing the moral graph, triangulated graph and the junction tree. Implementing message passing algorithm to get the cluster marginals.
Tools & Algorithms: Python 2, Bayesian Belief Networks