Man

Cristóbal Guzmán Paredes

Assistant Professor

Pontificia Universidad Católica de Chile

Santiago, Chile

Líneas de Investigación


Data Privacy; Convex Optimization; Analysis of Algorithms; Machine Learning

Educación

  •  Algorithms, Combinatorics and Optimization, GEORGIA INSTITUTE OF TECHNOLOGY. Estados Unidos, 2015
  •  Ingeniero Civil Matemático, UNIVERSIDAD DE CHILE. Chile, 2010

Experiencia Académica

  •   Assistant Professor Full Time

    PONTIFICIA UNIVERSIDAD CATOLICA DE CHILE

    Mathematics and Engineering

    Santiago, Chile

    2016 - At present

  •   Visiting Professor Full Time

    CENTRUM WISKUNDE & INFORMATICA

    Amsterdam, Holanda

    2017 - 2017

  •   Postdoctoral Researcher Full Time

    CENTRUM WISKUNDE & INFORMATICA

    Amsterdam, Holanda

    2015 - 2016

Experiencia Profesional

  •   Postdoctoral Researcher Full Time

    Universidad de Chile, Industrial Engineering Department

    Santiago, Chile

    2015 - 2015

  •   Lecturer: Engineering Optimization Part Time

    Georgia Institute of Technology, School of Industrial and Systems Engineering

    Atlanta, GA, Estados Unidos

    2015 - 2015

  •   Research Intern Full Time

    IBM, Almaden Research Center

    San José, CA, Estados Unidos

    2014 - 2014

  •   Research Assistant Part Time

    TU Delft, Institute of Applied Mathematics

    Delft, Holanda

    2012 - 2012

  •   Postdoctoral Researcher Full Time

    Centrum Wiskunde & Informatica

    Amsterdam, Holanda

    2015 - 2016

Formación de Capital Humano


Ph. D. Students:

[2021-2024] Clément Lezane, Non-Euclidean Stochastic Convex Optimization,
Ph.D. Applied Mathematics, University of Twente (in course)

M.Sc. Students:

[2021-2022] Tomás González, The Complexity of Differentially Private Stationary Points in Stochastic Nonconvex Optimization,
M.Sc. Engineering Universidad Católica (in course)

[2021-2022] Giorgio Micali, Membership Inference Attacks and Synthetic Data Generation with Differential Privacy,
M.Sc. Applied Mathematics, University of Twente

[2019-2022] Juan Pablo Flores, Stability and Privacy in Non-Euclidean Stochastic Optimization Methods,
M.Sc. Mathematics Universidad Católica (in course).

[2019-2020] Santiago Armstrong, Reconstructing Latent Filamentary Structure from Pairwise Correlations,
M.Sc. Engineering, Universidad Católica.

[2019-2021] Patricio Ulloa, Stability and Generalization of Stochastic Gradient Descent, with Applications in Differentially Private Data Analysis,
M.Sc. Engineering, Universidad Católica.

[2018-2020] Javiera Riffo, Dynamic Stackelberg Inspection Games,
M.Sc. Engineering, Universidad Católica.


Difusión y Transferencia


Industry Projects:

1. Director: Exploratory Data Analysis of the Production Process at Rosario Plant (with Agrosuper), 2018-2019


Premios y Distinciones

  •   Best Teaching Assistant

    UNIVERSIDAD DE CHILE

    Chile, 2008

    Awarded to the TA with highest scores in student teaching evaluations

  •   NIPS 2018 Top Reviewer

    Neural Information Processing Systems Foundation

    Canada, 2018

    Free registration to the Neural Information Processing Systems conference, awarded to the top 200 best reviewers (as scored by Area Chairs).

  •   Simons Research Visitor

    Simons Institute for the Theory of Computing

    Estados Unidos, 2019

    Research Visitor Scholarship for a long-term participation on the "Data Privacy: Foundations and Applications" Spring 2019 Semester.

  •   NeurIPS Top Reviewer

    Neural Information Processing Systems

    Canada, 2019

    Free registration for the NeurIPS 2019 conference, awarded to the top 400 reviewers

  •   NeurIPS spotlight paper

    Neural Information Processing Systems

    Estados Unidos, 2020

    Distinction given to the top 3% papers accepted at the conference. These papers are allowed to give a 10 minutes presentation of the paper, additionally to the publication in the conference proceedings and a poster presentation.


 

Article (10)

A sequential Stackelberg game for dynamic inspection problems
An Optimal Algorithm for Strict Circular Seriation
Statistical Query Algorithms for Mean Vector Estimation and Stochastic Convex Optimization
Lower Bounds for Parallel and Randomized Convex Optimization
Fast, deterministic and sparse dimensionality reduction
OPTIMAL AFFINE-INVARIANT SMOOTH MINIMIZATION ALGORITHMS
Lower Bounds on the Oracle Complexity of Nonsmooth Convex Optimization via Information Theory
New Upper Bounds for the Density of Translative Packings of Three-Dimensional Convex Bodies with Tetrahedral Symmetry
On Lower Complexity Bounds for Large-Scale Smooth Convex Optimization
Network congestion control with Markovian multipath routing

ConferencePaper (6)

Best-case lower bounds in online learning
Differentially Private Stochastic Optimization: New Results in Convex and Non-Convex Settings
Non-Euclidean Differentially Private Stochastic Convex Optimization
The complexity of nonconvex-strongly-concave minimax optimization
Stability of Stochastic Gradient Descent on Nonsmooth Convex Losses
Network pricing: How to induce optimal flows under strategic link operators

Proyecto (7)

Algorithmic Stability and Differential Privacy in Optimization, Machine Learning and Market Equilibria
First-Order Accelerated Methods in optimization and machine learning (FOAM)
Millenium Nucleus Center for the Discovery of Structures in Complex Data
EXPANDING THE SCOPE OF FIRST-ORDER METHODS IN OPTIMIZATION, LEARNING AND BEYOND
Limits of Linear Programming
Design of Efficient Saddle Point Algorithms for Large-scale/Complex Geometry Convex Optimization
ADAPTIVE DYNAMICS AND EQUILIBRIUM IN NETWORK FLOW GAMES
17
Cristóbal Guzmán

Assistant Professor

Mathematical Engineering

Pontificia Universidad Católica de Chile

Santiago, Chile

1
Jose Correa

Full Professor

Industrial Engineering

Universidad de Chile

Santiago, Chile

1
Roberto Cominetti

Full Professor

UNIVERSIDAD ADOLFO IBÁÑEZ

Santiago, Chile