Kiran Vodrahalli

{about} {research} {talks} {blog} {archive} {teaching} {notes} {links} {press}

talks


Research Talks

May 2022: Invited Talk at Meta Research.
The Platform Design Problem.

May 2022: Invited Talk at Google Brain AutoML.
Nonlinear Initialization Methods for Low-Rank Neural Networks.

May 2022: Invited Talk at Berkeley Center for Human-Compatible Artificial Intelligence (CHAI) Seminar.
Controllable and Interpretable Agents: The Platform Design Problem and Beyond.

April 2022: Invited Talk at Google Research NYC.
Sparse and Low-Rank: Resource-Efficient Methods in Machine Learning.

February 2022: Invited Talk at Google Brain Neural Modeling Group Invited Talk.
Nonlinear Initialization Methods for Low-Rank Neural Networks.

February 2022: Invited Talk at Simons Flatiron Center for Computational Neuroscience.
Sparse and Low-Rank: Resource-Efficient Methods in Machine Learning.

February 2022: Invited Talk at Amazon AWS.
Nonlinear Initialization Methods for Low-Rank Neural Networks.

February 2022: Invited Talk at Simons Theory of Computing, Learning in Games Program, Equilibrium Computation and Machine Learning Reading Group.
The Platform Design Problem.

February 2022: Graduating Bits at ITCS.
Graduating Bits Talk.
[slides-ppt] [slides-pdf] [YouTube video]

January 2022: Invited Talk at Simons Flatiron Center for Computational Mathematics.
Sparse and Low-Rank: Resource-Efficient Methods in Machine Learning.
[slides-ppt] [slides-pdf]

December 2021: NeurIPS 2021 StratML Workshop Spotlight Oral Presentation.
The Platform Design Problem.
[slides-ppt] [slides-pdf] [video]

December 2021: WINE 2021 Oral Presentation.
The Platform Design Problem.
[slides-ppt] [slides-pdf]

November 2021: Google Algorithms Group Invited Talk.
The Platform Design Problem.
[slides-ppt] [slides-pdf] [YouTube video]

October 2021: Google Learning Theory Group Invited Talk.
The Platform Design Problem.
[slides-ppt] [slides-pdf]

July 2021: EC 2021 NetEcon Workshop Oral Presentation.
The Platform Design Problem.
[workshop] [slides-ppt] [slides-pdf]

July 2021: ICML 2021 Long Oral Presentation (presented by Brandon Araki).
The Logical Options Framework.
[icml] [slides-pdf]

March 2020: NYAS 14th Annual Machine Learning Symposium Spotlight Presentation.
Learning the Optimal Step Size for Gradient Descent on Convex Quadratics.
[symposium] [slides-ppt] [slides-pdf]

October 2019: SOSP 2019 Oral Presentation (presented by Mathias Lécuyer).
Privacy Accounting and Quality Control in the Sage Differentially Private ML Platform.
[slides]

August 2019: Yahoo Research Invited Talk.
Attribute-Efficient Learning of Monomials over Highly-Correlated Variables.
[slides-ppt] [slides-pdf]

June 2019: RSS 2019 Oral Presentation (presented by Brandon Araki).
Learning to Plan with Logical Automata.
[slides] [slide-notes]

March 2019: ALT 2019 Oral Presentation.
Attribute-Efficient Learning of Monomials over Highly-Correlated Variables.
[slides-ppt] [slides-pdf]

March 2019: NYAS 13th Annual Machine Learning Symposium Spotlight Presentation.
Attribute-Efficient Learning of Monomials over Highly-Correlated Variables.
[symposium] [slides]

December 2018: NeurIPS 2018 Workshop on Infer2Control Oral Presentation (presented by Brandon Araki).
Learning to Plan with Logical Automata.
[slides]

July 2018: ICML 2018 Theory of Deep Learning Workshop Oral Presentation (presented by Mikhail Khodak).
A Compressed Sensing View of Unsupervised Text Embeddings, Bag-of-n-Grams, and LSTMs.

May 2018: Master’s Thesis presentation by Mikhail Khodak and Nikunj Saunshi.
A Compressed Sensing View of Unsupervised Text Embeddings, Bag-of-n-Grams, and LSTMs.
[slides]

April 2018: Seminar on Theoretical Machine Learning at the Institute for Advanced Study (presented by Mikhail Khodak).
A Compressed Sensing View of Unsupervised Text Embeddings, Bag-of-n-Grams, and LSTMs.
[slides]

September 2017: Invited talk at Princeton Neuroscience Institute.
Mapping between fMRI responses to movies and their natural language annotations.
[slides]

May 2017: Princeton COS Master’s Thesis Presentation.
Temporally Dependent Mappings Between fMRI Responses and Natural Language Descriptions of Natural Stimuli.
[slides]

December 2016: NeurIPS 2016 MLINI Workshop.
Mapping between Natural Movie fMRI Responses and Word-Sequence Representations.
[slides]

July 2016: Invited Talk at Intel-Princeton Neuroscience Collaboration Meeting.
Decoding fMRI to Text with Context.
[slides]

June 2016: ICML 2016 MVRL Workshop.
A Semantic Shared Response Model.
[slides]


Slides for Expository Presentations

There are several talks I’ve given in seminars and other settings which do not have slides — either they can be found on the notes page or I have not transcribed my handwritten notes into LaTeX yet. The following are the non-research presentations I’ve given with slides.