Welcome toMichael Psenka

The centralized location for everything Michael Psenka that no one asked for.

Download Curriculum Vitae

About Me

I am a second-year PhD student in EECS at UC Berkeley, advised by Prof. Yi Ma and Prof. Shankar Sastry. I work at the intersection of geometry and deep learning, developing theory and currently using this theory for applications in 3D vision and robotics.

Theoretically, I employ mathematical techniques from modern geometric fields (e.g. Riemannian geometry) to study study any kind of data, not just vision data; I then apply this theory by developing robust AI suitable for industry use.

Name : Michael P. Psenka
Date of birth : June 10, 1998
Email : psenka@berkeley.edu
Phone : +1 (843) 530 3503

Education

University of California, Berkeley | PhD in Electrical Engineering and Computer Science, focus on AI and robotics

Sept 2021 - Jun 2026

Princeton Univeristy | A.B. Math, certificates in Applied and Computational Mathematics, Applications of Computing

Sept 2017 - Jun 2021

Employment

Co-Head Instructor @ UC Berkeley

Jun 2022 - Aug 2022
  • Organized and taught lectures for CS 70, an undergraduate class for discrete math and probability theory
  • Link to class page.

Undergrad Researcher @ Stanford University

Jun 2020 - Aug 2020
  • Worked with Dr. Tolga Birdal on a novel approach to multi-view reconstruction in computer vision that bypasses pairwise view registration

Undergrad Researcher @ Princeton University

Jun 2019 - Sept 2019
  • Undergraduate research funded by the National Science Foundation through award DMS-1719558
  • Successfully developed a state-of-the-art method for computing analytic Hessians and second order optimization over tensor train manifolds

Software Developer @ Moovila

June - Aug 2018, '17, '16
  • Developed a machine learning algorithm for workplace analytics
  • Mathematically modeled collision avoidance in network analysis animation
  • Worked through a patent application for proprietary software
  • Worked on improving the search engine for quicker and more robust search results
  • Denormalized relational database to NoSQL, maximizing data access efficiency and cost-efficiency

Awards

Peter A. Greenberg ’77 Memorial Prize

June 2020
  • Awarded for outstanding accomplishments in Mathematics by juniors

Manfred Pyka Memorial Prize

June 2018
  • Given to outstanding Physics undergraduates who have shown excellence in course work and promise in independent research

HackPrinceton First Place

April 2018
  • Won first place for developing A.I.D.A.N. at HackPrinceton Spring 2018. Link to project.
  • A.I.D.A.N. is a chatbot that, once you upload your dataset, has the capacity to do useful statistical analysis and machine learning on your dataset through voice commands

I focus on representation learning, and how we can find ideal represenetations of data automatically and explicitly.

Deep learning has shown impressive power in learning compact representations of otherwise complex data, such as images or natural language. While geometric methods like contrastive learning have proven to be beneficial for learning good representations, these methods are often impractical at scale and are traded off for hand-engineered hyperparameter tuning, data augmentation etc.

I aim to modernize Riemannian geometry for a computational world, so that geometric-based representation learning methods like contrastive learning can scale in the way modern computing demands. While I currently focus on computer vision and robotics, I believe the "geometric viewpoint" of representation learning can provide both interpretability and power to any area that a neural network may be useful.

Publications & Workshop Presentations

Michael Psenka, Druv Pai, Vishal Raman, Shankar Sastry, Yi Ma. Representation Learning through Manifold Flattening and Reconstruction submitted to SLowDNN. Link to paper.

Nov 2022

Druv Pai, Michael Psenka, Chih-Yuan Chiu, Manxi Wu, Edgar Dobriban, Yi Ma. Pursuit of a discriminative representation for multiple subspaces via sequential games submitted to Journal of the Franklin Institute. Link to paper.

Sept 2022

Xili Dai, Shengbang Tong, Mingyang Li, Ziyang Wu, Michael Psenka, Kwan Ho Ryan Chan, Pengyuan Zhai, Yaodong Yu, Xiaojun Yuan, Heung Yeung Shum, Yi Ma. CTRL: Closed-Loop Transcription to an LDR via Minimaxing Rate Reduction Published in Entropy. Link to paper.

Nov 2021

Michael Psenka and Nicolas Boumal. Second-order optimization for tensors with fixed tensor-train rank. Poster presentation at OPT2020. Link to paper. Link to poster.

Dec 2020

Michael Psenka, Tolga Birdal, Leonidas Guibas. Reconstruction Without Registration. Video presentation at IROS2020 geometric methods workshop. Link to paper. Link to presentation.

Oct 2020

Ryan Arbon, Mohammed Mannan, Michael Psenka, Seyoon Ragavan. A Proof of The Triangular Ashbaugh-Benguria-Payne-Pólya-Weinberger Inequality. To appear in Journal of Spectral Theory. Link to paper.

Sept 2020

Please see my CV for more research projects and further details.

My Music

I'm a classically trained pianist with some experience playing jazz. Recently, I started tinkering around with making music in a digital workstation. Here, I'll upload anything that becomes a finished product.

I'm also a member of the Princeton Pianist Ensemble! My most recent performance was a duet version of Fly Me To The Moon for a virtual concert: Link to video. You can find more from the ensemble here.

1. Hurricane

Drum sample: Incredible Bongo Band - Apache

2. Deadly

Drum sample: Kanye West - Black Skinhead

Get in touch

Email

psenka@berkeley.edu

Close