Juan Elenter

Ph.D. Candidate
University of Pennsylvania

elenter [AT] seas.upenn.edu

Bio

> I'm a 3rd year Ph.D. student at the University of Pennsylvania, where I mostly think about constrained optimization, LLM finetuning, active learning and continual learning with Prof. Ribeiro. I'm also working towards a Masters degree in Data Science and Statistics from The Wharton School . Before graduate school, I visited Anshul Kundaje at Stanford's Kundaje-Lab, where I worked on genomic motif discovery through ATAC-seq analysis. I also worked at CERN, contributing to the CMS Open Data initiative with Dr. Lassila-Perini.

I'm originally from the beautiful city of Montevideo, Uruguay, where I obtained a BSc. in Electrical Engineering from UdelaR. During my undergrad, I was a software developer at IBM . If any of this sparks your interest, or if you want share some Yerba Mate , feel free to reach out.

My work

On the theory side, I work on duality-based constrained optimization, which is particularly relevant in the context of LLM fine tuning and foundation models. This endeavour is about obtaining models that not only excel at a main task, but also adhere to requirements such as safety, invariance, robustness and fairness. For instance, my last paper shows that, in many learning problems, dual subgradient methods yield near-feasible solutions, without randomization, despite non-convexity.

On the applications side, I have delved into data-centric ML and explored the following questions:
- Which set of samples should I label next so that my model is the best possible with the lowest labelling cost ?
- What are the most impactful samples in my dataset? Which ones should I discard ?
- How should I continuously finetune my model as new data, with different properties, gets collected ?
These interrogations have led to works on active and continual learning.

I have a particular interest in problems involving biological signals such as Genome Sequences , Medical Images, Brain Images and the Gut Microbiome.

Near-Optimal Solutions of Constrained Learning Problems.

Juan Elenter, Luiz Chamon, Alejandro Ribeiro

International Conference on Learning Representations (ICLR), 2024

Primal-Dual Continual Learning: Stability and Plasticity through Lagrange Multipliers.

Juan Elenter, Navid NaderiAlizadeh, Tara Javidi, Alejandro Ribeiro

Preprint Under Review

A Lagrangian Duality Approach to Active Learning

Juan Elenter, Navid NaderiAlizadeh, Alejandro Ribeiro

Neural Information Processing Systems, 2022.

Neural Networks with Quantization Constraints.

Ignacio Hounie*, Juan Elenter*, Alejandro Ribeiro

IEEE International Conference on Acoustics, Speech, and Signal Processing, 2023.

Graph Neural Networks for genome enabled prediction of complex traits.

Juan Elenter, Ignacio Hounie, Guillermo Etchebarne, María Inés Fariello, Federico Lecumberry

Probabilistic Modeling in Genomics, CSHL, 2021.

Projects

ATAC-seq analysis and motif discovery
@ Stanford, KundajeLab
Undergrad Capstone Project: Prediction of complex traits in agriculture.
@ UdelaR, FarielBerry Lab
I find this case very confusing, and have not thoroughly checked the result. R.A.Fisher
CMS Open Data Initiative
@ CERN, CMS

Vitæ

  • Flatiron Institute, Simons Foundation 2022
    Summer Intern
    Center for Computational Mathematics
  • University of Pennsylvania Fall 2021 - now
    Ph.D. Candidate
    Optimization theory, Signal Processing, Machine Learning
  • Stanford University Spring 2021
    Visiting Student at KundajeLab
    ATAC-seq analysis, motif discovery
  • CERN 2020
    Intern
    Compact Muon Solenoid Experiment
  • IBM 2018-2019
    Full Stack Developer
  • ProteanTecs Summer 2019
    Signal Processing Intern
    Chip Predictive Maintenance
  • Universidad de la República 2016-2021
    Bsc. Electrical Engineering
    Signal Processing

Miscelaneous

  • Uruguay
  • Water

Some stuff I like about Uruguay.

Mate: 1/2 Moncayo + 1/2 Baldo
My only real conversation is with this green gourd. Julio Cortázar
The uruguayan
We're kind of chill, obsessed with soccer, skeptical of authority and fixated on social welfare.
La Negra Tomasa
The best pizza in Uruguay, provably.
Jorge Drexler
No le hagas caso a tanto misterio
vos ya sabés la verdad
que no hay nada peor para esta seriedad
que tomársela en serio.
Asado
Typical sunday meal at family gatherings.
Murga
Asaltantes Con Patentes 2013

Water is my natural environment.


Involuntarily drinking water in Punta Ballena.

Glassy morning in one of my favourite uruguayan spots.

Involuntarily drinking frozen, cristalyzed water.

West Santa Lucia River by the way.

Me and my uncle Ernesto, with whom I crossed the Atlantic in 2016.

In 2013, while playing for Uruguay's U-17 national team I lost 3-23 to the US.
The blueprint for this website can be found in this GitHub repo. Feel free to use it.