Christopher Waites

Computer science student @
­čî▓Stanford University ­čî▓

Broadly interested in deep learning,
generative modeling, privacy,
and adversarial ML.

GitHub Profile

Projects

JAX-Flows: Normalizing Flows in JAX

Differentially Private Mixed-Type Data Generation

PyVacy: Differentially Private Optimization Algorithms for PyTorch

Reversible Flow Models for Improved Biometric Verification Systems

Blog

When Differential Privacy Might Be Most Useful

An Introduction to Differentially Private Deep Learning

Teaching

CS231n: Convolutional Neural Networks for Visual Recognition
Stanford University, Spring 2020

CS230: Deep Learning
Stanford University, Winter 2020

CS221: Artificial Intelligence
Stanford University, Autumn 2019

CS7646: Machine Learning for Trading
Georgia Institute of Technology, Autumn 2018

CS3600: Artificial Intelligence
Georgia Institute of Technology, Autumn 2017

CS1331: Object-Oriented Programming
Georgia Institute of Technology, Summer 2016

Talks

Privacy-Preserving Deep Learning
Definition of differential privacy, the Laplace mechanism, differentially private stochastic gradient descent (DP-SGD), and PATE.
CS271: AI in Healthcare, Stanford University, Winter 2020.

Differentially Private Synthetic Data Generation
Definition of differential privacy, differentially private stochastic gradient descent (DP-SGD), and generative adversarial networks.
Two Sigma Investments, Summer 2019.

Introductory Topics in Theoretical Computer Science
Computability, deterministic and nondeterministic finite automata, regular languages, the pumping lemma, and Turing machines.
Facebook, Inc., Spring 2018.

What IÔÇÖm Reading

AutoML-Zero: Evolving Machine Learning Algorithms From Scratch

Adversarial Examples Are Not Bugs, They Are Features

On the Measure of Intelligence

Privacy and Synthetic Datasets

Keeping Top AI Talent in the United States

Deep Learning for Symbolic Mathematics

Building Machines That Learn and Think Like People

Deep Image Reconstruction from Human Brain Activity

The Linear Algebra Mapping Problem

Neural Turing Machines

Design