Research
I am interested in developing algorithms for reliable and trustworthy machine learning, with a particular focus on representation learning, selfsupervision and continual learning.


Efficient Parametric Approximations of Neural Network Function Space Distance
Nikita Dhawan,
Sicong (Sheldon) Huang,
Juhan Bae,
Roger Grosse
ICML, 2023
arXiv
We consider a specific case of approximating the function space distance (FSD) over the
training set, i.e. the average distance between the outputs of two ReLU neural networks, based on approximating the architecture as a
linear network with stochastic gating. Despite requiring only one
parameter per unit of the network, our parametric approximation is competitive with stateoftheart nonparametric approximations
with larger memory requirements, when applied to continual learning and influence function estimation.


Dataset Inference for SelfSupervised Models
Adam Dziedzic,
Haonan Duan,
Muhammad Ahmad Kaleem,
Nikita Dhawan,
Jonas Guan,
Yannis Cattan,
Franziska Boenisch,
Nicolas Papernot
NeurIPS, 2022
arXiv
We introduce a new dataset inference defense for selfsupervised models, which uses the intuition that the loglikelihood of an encoder's output representations is higher on the victim's training data than
on test data if it is stolen from the victim, but not if it is independently trained. Our extensive empirical results in the vision
domain demonstrate that dataset inference is a promising direction for defending selfsupervised models against model stealing.


On the Difficulty of Defending SelfSupervised Learning against Model Extraction
Adam Dziedzic,
Nikita Dhawan,
Muhammad Ahmad Kaleem,
Jonas Guan,
Nicolas Papernot
ICML, 2022
arXiv
Recently, MLasaService providers have commenced offering trained selfsupervised models over inference APIs, which transform
user inputs into useful representations for a fee. However, the high cost involved to train these models and their exposure over APIs
both make blackbox extraction a realistic security threat. We explore model stealing by constructing several novel attacks and evaluating
existing classes of defenses.


ARM: A MetaLearning Approach for Tackling Group Shift
Marvin Zhang*,
Henrik Marklund*,
Nikita Dhawan*,
Abhishek Gupta,
Sergey Levine,
Chelsea Finn
NeurIPS, 2021
website /
arXiv
Machine learning systems are regularly tested under distribution shift, in reallife applications. In this work, we consider the setting where
the training data are structured into groups and test time shifts correspond to changes in the group distribution. We propose to use ideas from
metalearning to learn models that are adaptable, and introduce the framework of adaptive risk minimization (ARM), a formalization of this setting.


AVID: Learning MultiStage Tasks via PixelLevel Translation of Human Videos
Laura Smith,
Nikita Dhawan,
Marvin Zhang,
Pieter Abbeel,
Sergey Levine,
RSS, 2020
website /
arXiv /
blog
Humans can learn from watching others, imagining how they would perform the task themselves, and then practicing on their own.
Can robots do the same? We adopt a similar strategy of imagination and practice in this project to solve complex, longhorizon tasks,
like operating a coffee machine or getting objects from within a closed drawer.

Student Researcher
Google, April 2023  Present
Hosted by Nicole Mitchell and Karolina Dziugaite.


CSC 311: Introduction to Machine Learning Fall 2021 (University of Toronto)
EECS 126: Probability and Random Processes Fall 2020, Spring 2020 (UC Berkeley)
EECS 229A: Information Theory and Coding Fall 2020 (UC Berkeley)

