I'm a PhD student in Computer Science at the University of Toronto and Vector Institute.
My research interests lie broadly in machine learning, with a focus on generative models.
Last summer, I worked as a machine learning intern at Microsoft Research.

My PhD advisors are Quaid Morris and David Duvenaud.

I'm currently co-organizing a workshop at ICML 2019 on Invertible Neural Networks and Normalizing Flows.

On the Importance of Learning Aggregate Posteriors in
Multimodal Variational Autoencoders
We study latent variable models of two modalities: images and text. A common task for these multimodal models is to perform conditional generation; for instance, generating an image conditioned on text. This can be achieved by sampling the posterior of the text then generating the image given the latent variable. However, we find that a problem with this approach is that the posterior of the text does not match the posteriors of the images corresponding to that text. The result is that the generated images are either of poor quality or don't match the text. A similar problem is also encountered in the mismatch between the prior and the marginal aggregate posterior. In this paper, we highlight the importance of learning aggregate posteriors when faced with these types of distribution mismatches. We demonstrate this on modified versions of the CLEVR and CelebA datasets. |
||

Inference Suboptimality in Variational Autoencoders
We analyze approximate inference in variational autoencoders in terms of the approximation and amortization gaps. We find that suboptimal inference is often due to amortizing inference rather than the limited complexity of the approximating distribution. We show that this is due partly to the generator learning to accommodate the choice of approximation. Furthermore, we show that the parameters used to increase the expressiveness of the approximation play a role in generalizing inference rather than simply improving the complexity of the approximation. |
||

Reinterpreting Importance-Weighted Autoencoders
The standard interpretation of importance-weighted autoencoders is that they maximize a tighter lower bound on the marginal likelihood than the standard evidence lower bound. We give an alternate interpretation of this procedure: that it optimizes the standard variational lower bound, but using a more complex distribution. We formally derive this result, present a tighter lower bound, and visualize the implicit importance-weighted distribution. |

Learning to Ignore
An exploration of how to model information that is relevant to a trained network |
||

Uncertainty in Bayesian Neural Networks
Visualizations of decision boundaries in BNNs |
||

Approximate Posterior Building Blocks
This is a short review of orthogonal methods for improving inference in latent variable models. I examine the lower bounds and complexities of these models as well as their combinations. |
||

Gradients of Deep Networks
A small look at skip-connection models |
||

Intro to Probability for ML
Review of probability theory |

MSc Dissertation: Gene Expression Deconvolution with Subpopulation Proportions
Personalized cancer strategies are currently being hindered by intratumor heterogeneity.
One source of heterogeneity, clonal evolution, can lead to genetically distinct subpopulations
within a sample. Through the use of subclonal reconstruction methods, we can obtain estimates
of the subpopulation proportions within a single sample. Here, I leverage these proportion
estimates by incorporating it into the deconvolution of tumour gene expression data in order to
estimate the subclone specific gene expression profiles. |
||

DeepVec
As a side project, I started a deep learning company that helps businesses automate their workflow and learn from their data. We are currently serving a driving safety company where we automate the classification of their driving videos for their driver safety review process. |