cv

PhD

Department of Computer Science, University of Bath

Accountable, Responsible, and Transparent AI

MSc

Department of Computer Science, University of Bath

Accountable, Responsible, and Transparent AI

• Bayesian Machine Learning
• Robotics Software
• AI as a Social and Political Practice

MSc

School of Electrical Engineering, University of Belgrade

Applied Mathematics

• Machine Learning
• Artificial Intelligence
• Probability and Statistics

BSc

School of Electrical Engineering, University of Belgrade

Computer Engineering and Informatics

• Linear Algebra
• Calculus
• Numerical and Discrete Mathematics
• Probability and Statistics
• Artificial Intelligence
• Operating Systems
• Compilers
• Object-Oriented Programming
• Concurrent and Distributive Programming
• Software Design
• Cryptography

Research Engineer

Incode

Developed machine learning models that prevented facial spoofing and enabled user verification.

• image classification
• outlier detection
• facial spoofing
• face liveness
• computer vision

Research Scientist

Everseen

Developed computer vision machine learning models that verified the self-checkout process in retail stores.

• image classification
• object detection
• expert systems
• computer vision

Research Engineer

Retail Intelligence LLC

Developed computer vision machine learning models that verified the checkout process in retail stores.

• image classification
• object detection
• action recognition
• pose estimation
• computer vision

Unit Convenor

Practical Seminar in Machine Learning

Unit convenor for Reinforcement Learning, with occasional involvement in summer school organization.

Intrinsically motivated option learning: a comparative study of recent methods

IEEE

Options represent a framework for reasoning across multiple time scales in reinforcement learning (RL). With the recent active interest in the unsupervised learning paradigm in the RL research community, the option framework was adapted to utilize the concept of empowerment, which corresponds to the amount of influence the agent has on the environment and it's ability to perceive this influence, and which can be optimized without any supervision provided by the environment's reward structure. Many recent papers modify this concept in various ways achieving commendable results. Through these various modifications, however, the initial context of empowerment is often lost. In this work we offer a comparative study of such papers through the lens of the original empowerment principle.