Scholar
Aleksandar Petrov
Google Scholar ID: em54BT4AAAAJ
Research Scientist, DeepMind
Citations & Impact
Citations
674
H-index
9
i10-index
9
Publications
20
Co-authors
0
Publications
Resume
Academic Achievements
- - Publications: Multiple papers accepted at top conferences such as ICLR, NeurIPS, and ICML
- - Awards: Entropic Award for most surprising negative result at the I Can't Believe It's Not Better workshop at NeurIPS 2023
- - Projects: Developed a programming language for recurrent models
- - Contributed to the Alan Turing Institute’s response to the House of Lords Large Language Models Call for Evidence
Research Experience
- - Google DeepMind: Research Scientist, part of the SynthID team
- - Motional: Research Intern
- - Adobe: Research Intern
- - Meta: Research Intern
- - FAIR: Research Intern, working on watermarking
- - University of Oxford: Ph.D. research, supervised by Philip Torr and Adel Bibi
Education
- - Ph.D.: University of Oxford, Autonomous Intelligent Machines and Systems CDT, supervised by Philip Torr and Adel Bibi
- - M.Sc.: ETH Zürich, focusing on robotics, machine learning, statistics, and applied category theory. Thesis: Compositional Computational Systems
- - During his time at ETH, he worked closely with Prof. Emilio Frazzoli's group and was generously funded by the Excellence Scholarship & Opportunity Programme (ESOP).
Background
- - Research Interests: Understanding the fundamental properties of deep learning systems to make them more reliable, robust, efficient, and broadly beneficial for society.
- - Professional Field: Multimedia provenance and watermarking
- - Brief Introduction: Currently working at Google DeepMind on multimedia provenance and watermarking as part of the SynthID effort.