Walid Krichene

I am a research scientist at Google Research, where I work on optimization, differential privacy, and recommender systems. I also work on developing optimization methods using continuous-time and stochastic dynamics. I co-authored the Recommendation Systems class of Google's open-source ML courses, and I teach at the ML@CapitalG.

I did my Ph.D. in Electrical Engineering and Computer Sciences at UC Berkeley, where I was advised by Alex Bayen and Peter Bartlett. My thesis was on Continuous and discrete time dynamics for online learning and convex optimization.

News

News Archive

Publications

2024

  • W. Krichene, N. Mayoraz, S. Rendle, S. Song, A. Thakurta, and L. Zhang. Private Learning with Public Features, AISTATS 2024.
    bibtex abstract

2023

  • M Sundararajan and W. Krichene. Inflow, Outflow, and Reciprocity in Machine Learning, ICML 2023 (oral).
    bibtex abstract pdf Oral
  • W. Krichene, P. Jain, S. Song, M, Sundararajan, A. Thakurta, and L. Zhang. Multi-task Differential Privacy under Distribution Skew, ICML 2023.
    bibtex abstract pdf Talk
  • M. Curmei, W. Krichene, L. Zhang, and M. Sundararajan. Private Matrix Factorization with Public Item Features, RecSys 2023.
    bibtex abstract pdf

2022

  • H. Mehta, W. Krichene, A. Thakurta, A. Kurakin, and A. Cutkosky Differentially private image classification from features, Transactions on Machine Learning Research 2022.
    bibtex abstract pdf Blog
  • S. Rendle, W. Krichene, L. Zhang, and Y. Koren. Revisiting the performance of ials on item recommendation benchmarks, RecSys 2022.
    bibtex abstract pdf

2021

  • S. Chien, P. Jain, W. Krichene*, S. Rendle, S. Song, A. Thakurta* and L. Zhang. Private Alternating Least Squares: Practical Private Matrix Completion with Tighter Rates, ICML 2021 (long oral).
    bibtex abstract pdf Talk
  • W. Krichene and S. Rendle. On Sampled Metrics for Item Recommendation (Extended Abstract), IJCAI 2021 (Sister Conferences Best Papers).
    pdf abstract

2020

  • W. Kong*, W. Krichene*, N. Mayoraz, S. Rendle and L. Zhang. Rankmax: An Adaptive Projection Alternative to the Softmax Function, NeurIPS 2020.
    bibtex abstract pdf Talk
  • W. Krichene, K. F. Caluya and A. Halder. Global Convergence of Second-order Dynamics in Two-layer Neural Networks, preprint.
    bibtex abstract pdf
  • W. Krichene and S. Rendle. On Sampled Metrics for Item Recommendation, KDD 2020. Best Paper Award
    bibtex abstract pdf Talk
  • S. Rendle, W. Krichene, L. Zhang and J. Anderson. Neural Collaborative Filtering vs. Matrix Factorization Revisited, RecSys 2020.
    bibtex abstract pdf Talk slides
  • J. Anderson, Q. Huang, W. Krichene, S. Rendle and L. Zhang. Superbloom: Bloom filter meets Transformer, 2020.
    bibtex abstract pdf

2019

  • F. Belletti, K. Lakshmanan, W. Krichene, N. Mayoraz, Y. Chen, J. Anderson, T. Robie, T. Oguntebi, D. Shirron and A. Bleiwess. Scaling Up Collaborative Filtering Data Sets through Randomized Fractal Expansions, 2019.
    bibtex abstract pdf
  • W. Krichene, N. Mayoraz, S. Rendle, L. Zhang, X. Yi, L. Hong, E. Chi and J. Anderson. Efficient Training on Very Large Corpora via Gramian Estimation, ICLR 2019.
    bibtex abstract pdf poster

2018

  • W. Krichene, M. C. Bourguiba, K. Lam, and A. Bayen. On Learning How Players Learn: Estimation of Learning Dynamics in the Routing Game. ACM Transactions on Cyber-Physical Systems - Special Issue.
    bibtex abstract pdf
  • S. Samaranayake, W. Krichene, J. Reilly, M. Delle Monache, P. Goatin, and A. Bayen. Discrete-Time System Optimal Dynamic Traffic Assignment (SO-DTA) with Partial Control for Physical Queuing Networks.. Transportation Science.
    bibtex abstract

2017

  • W. Krichene and P. Bartlett. Acceleration and Averaging in Stochastic Descent Dynamics. 31st Annual Conference on Neural Information Processing Systems (NIPS).
    bibtex abstract pdf supplement arxiv spotlight
  • W. Krichene, J. D. Reilly, S. Amin, and A. Bayen. Stackelberg Routing on Parallel Transportation Networks. Book Chapter. In: Basar T., Zaccour G. (eds) Handbook of Dynamic Game Theory. Springer.
    bibtex abstract pdf

2016

  • W. Krichene, A. Bayen and P. Bartlett. Adaptive Averaging in Accelerated Descent Dynamics. 30th Annual Conference on Neural Information Processing Systems (NIPS).
    bibtex abstract pdf supplement spotlight
  • M. Balandat, W. Krichene, A. Bayen and C. Tomlin. Minimizing regret on reflexive Banach spaces and Nash equilibria in continuous zero-sum games. 30th Annual Conference on Neural Information Processing Systems (NIPS).
    bibtex abstract pdf supplement
  • W. Krichene, M. Suarez Castillo, and A. Bayen. On Social Optimal Routing Under Selfish Learning. Transactions on Contol of Network Systems (TCNS).
    bibtex abstract pdf

2015

  • W. Krichene, A. Bayen, P. Bartlett. Accelerated Mirror Descent in Continuous and Discrete Time. 29th Annual Conference on Neural Information Processing Systems (NIPS).
    bibtex abstract pdf supplement code
  • W. Krichene, S. Krichene, and A. Bayen. Efficient Bregman Projections onto the Simplex. 54th IEEE Conference on Decision and Control (CDC).
    bibtex abstract pdf code
  • W. Krichene, B. Drighes, and A. Bayen. Online Learning of Nash Equilibria in Congestion Games. SIAM Journal on Control and Optimization, 53(2), 1056–1081.
    bibtex abstract arxiv
  • S. Krichene, W. Krichene, R. Dong, and A. Bayen. Convergence of Heterogeneous Distributed Learning in Stochastic Routing Games. 53rd Annual Allerton Conference on Communication, Control, and Computing.
    bibtex abstract pdf
  • R. Dong, W. Krichene, A. Bayen, and S. Sastry. Differential Privacy of Populations in the Routing Game. 54th IEEE Conference on Decision and Control (CDC).
    bibtex abstract pdf arxiv
  • W. Krichene, M. Balandat, C. Tomlin, and A. Bayen. The Hedge Algorithm on a Continuum. International Conference on Machine Learning (ICML).
    bibtex abstract pdf supplement talk
  • Z. Abrams Bayen, S. Chen, J. W. Frank, A. Gorajek, W. Krichene, and Itamar Rossen. Systems and Methods for Estimating User Attention (Patent No US20150142954 A1).
    pdf

2014

  • W. Krichene, B. Drighes, and A. Bayen. On the Convergence of No-Regret Learning in Selfish Routing. 31st International Conference on Machine Learning (ICML).
    bibtex abstract pdf supplement talk
  • B. Drighes, W. Krichene, and A. Bayen. Stability of Nash Equilibria in the Congestion Game under Replicator Dynamics. 53rd IEEE Conference on Decision and Control (CDC).
    bibtex abstract pdf
  • W. Krichene, J. Reilly, S. Amin, A. Bayen. Stackelberg Routing on Parallel Networks with Horizontal Queues. IEEE Transactions on Automatic Control (TAC).
    bibtex abstract pdf
  • J. Reilly, S. Samaranayake, M. Delle Monache, W. Krichene, P. Goatin, and A. Bayen. Adjoint-based optimization on a network of discretized scalar conservation law PDEs with applications to coordinated ramp metering. Journal of Optimization Theory and Applications (JOTA).
    bibtex abstract pdf
  • M. Delle Monache, J. Reilly, S. Samaranayake, W. Krichene, P. Goatin, and A. Bayen. A PDE-ODE Model for a Junction with Ramp Buffer. SIAM Journal on Applied Mathematics.
    bibtex abstract pdf

Thesis

  • W. Krichene. Continuous and Discrete Time Dynamics for Online Learning and Convex Optimization.
    bibtex abstract pdf

Some older talks

RecSys 2020
Neural Collaborative Filtering Vs Matrix Factorizaion Revisited.
Virtual Conference. Sep. 24, 2020.
video
KDD 2020
On Sampled Metrics for Item Recommendation.
Virtual Conference. Aug. 2020.
Google Research Conference

Acceleration and Averaging in Stochastic Descent Dynamics.
Mountain View, CA. Nov. 15, 2017.

NIPS 2017 Spotlight
Acceleration and Averaging in Stochastic Descent Dynamics.
Long Beach, CA. Dec. 5, 2017.
poster session video
NIPS 2016 Spotlight
Adaptive Averaging in Accelerated Descent Dynamics.
Barecelona, Spain. Dec. 2016.
poster
Dissertation talk. EECS, U.C. Berkeley.
Continuous and Discrete Dynamics for Online Learning and Convex Optimization.
Berkeley, CA. August 18, 2016.
slides
LIDS seminar, MIT

Online Learning and Optimization: From Continuous to Discrete.
Boston, MA. April 5, 2016.
slides

Microsoft Research
Online Learning and Optimization: From Continuous to Discrete Time.
New York, NY. Feb. 18, 2016.
NIPS 2015 Spotlight

Accelerated Mirror Descent in Continuous and Discrete Time.
Montreal, Canada. Dec. 2015.
poster

IEEE Conference on Decision and Control

Efficient Bregman Projections onto the Simplex.
Osaka, Japan. Dec. 16, 2015.
slides

ICML 2015

The Hedge Algorithm on a Continuum.
Lille, France. Jul. 8, 2015.
slides poster

ICML 2014

On the convergence of online learning in selfish routing.
Beijing, China. Jun. 23, 2014.
slides poster

Teaching

AMLD Africa
Summer 2021
Workshop on Recommendation Systems.
ML@CapitalG
2017-2021
Recommender Systems.
Nassma ML summer school
Summer 2019
Optimization Methods; Recommender Systems.
The Next MBA
2018-2019
Introduction to Machine Learning.
Berkeley MFE230P
Fall 2014
Optimization Methods for Finance.
Recipient of the Berkeley Outstanding Graduate Student Instructor Award (2015).
Berkeley EE128
Fall 2013
Feedback Control Systems.
Recipient of the EECS Outstanding Graduate Instructor Award (2014).
student project videos