Date/Time : 12 Nov 2021 10:00
Venue : via Zoom
Speaker : Clement Canonne, University of Sydney

Join Zoom Meeting
https://nus-sg.zoom.us/j/9775606531?pwd=cVZ6eTRYS3ZzVUlFNXRuSlhMVjJSUT09

Meeting ID: 977 560 6531
Passcode: 123321

Abstract: In this talk, I will introduce the primitive of “domain compression” over the probability simplex, which enables us to reduce the domain size from k to L while contracting (with high probability) the pairwise total variation distances by a factor at most √(L/k). This can equivalently be seen as some type of dimensionality reduction for probability distributions, preserving (in a one-sided way) the L1 distances.

Starting with the corresponding statement for L=2, I will show how to perform this domain compression in a randomness-efficient fashion, and discuss its optimality. As an application, I will describe how this yields sample-optimal algorithms for goodness-of-fit under various constraints, such as communication (bandwidth) constraints, local privacy, and shuffle privacy.

Based on joint works with Jayadev Acharya, Cody Freitag, Yanjun Han, Ziteng Sun, and Himanshu Tyagi (in particular, (https://arxiv.org/abs/1907.08743; COLT 2020)); and some insights from a paper by Amin, Joseph, and Mao (https://arxiv.org/abs/1911.01452; COLT 2020).

Add to calendar :

Date/Time : 12 Nov 2021 16:00
Venue : via Zoom
Speaker : Alexandros Beskos, University College of London

Join Zoom Meeting
https://nus-sg.zoom.us/j/9775606531?pwd=cVZ6eTRYS3ZzVUlFNXRuSlhMVjJSUT09

Meeting ID: 977 560 6531
Passcode: 123321

Abstract: Bayesian inference for nonlinear diffusions, observed at discrete times,is a challenging task that has prompted the development of a number of algorithms, mainly within the computational statistics community. We propose a new direction, and accompanying methodology – borrowing ideas from statistical physics and computational chemistry – for inferring the posterior distribution of latent diffusion paths and model parameters, given observations of the process. Joint configurations of the underlying process noise and of parameters, mapping onto diffusion paths consistent with observations, form an implicitly defined manifold. Then, by making use of a constrained Hamiltonian Monte Carlo algorithm on the embedded manifold, we are able to perform computationally efficient inference for a class of discretely observed diffusion models. Critically, in contrast with other approaches proposed in the literature, our methodology is highly automated, requiring minimal user intervention and applying alike in a range of settings, including: elliptic or hypo-elliptic systems; observations with or without noise; linear or non-linear observation operators. Exploiting Markovianity, we propose a variant of the method with complexity that scales linearly in the resolution of path discretisation and the number of observation times. Example Python code is given at git.io/m-mcmc.

Add to calendar :

Date/Time : 29 Oct 2021 10:00
Venue : via Zoom
Speaker : Jian Tang, Mila-Quebec Al Institute

Join Zoom Meeting
https://nus-sg.zoom.us/j/9775606531?pwd=cVZ6eTRYS3ZzVUlFNXRuSlhMVjJSUT09

Meeting ID: 977 560 6531
Passcode: 123321

Abstract:  Drug discovery is a very long and expensive process, taking on average more than 10 years and costing $2.5B to develop a new drug. Artificial intelligence has the potential to significantly accelerate the process of drug discovery by extracting evidence from a huge amount of biomedical data and hence revolutionizes the entire pharmaceutical industry. In particular, graph representation learning and geometric deep learning–a fast growing topic in the machine learning and data mining community focusing on deep learning for graph-structured and 3D data—has seen great opportunities for drug discovery as many data in the domain are represented as graphs or 3D structures (e.g. molecules, proteins, biomedical knowledge graphs). In this talk, I will introduce our recent progress on geometric deep learning for drug discovery and also a newly released open-source machine learning platform for drug discovery, called TorchDrug.

Add to calendar :

Date/Time : 28 Oct 2021 16:00
Venue : via Zoom
Speaker : Kunlun Qi, The Chinese University of Hong Kong

Join Zoom Meeting
https://nus-sg.zoom.us/j/81670327104?pwd=RnJwRm9rblhuYkRZS1dlY1NvOU91Zz09

Meeting ID: 816 7032 7104
Passcode: 779661

Abstract:  Numerical approximation of the Boltzmann equation is a challenging problem due to its high-dimensional, nonlocal, and nonlinear collision integral. Over the past decade, the Fourier-Galerkin spectral method has become a popular deterministic method for solving the Boltzmann equation, manifested by its high accuracy and potential of being further accelerated by the fast Fourier transform. Albeit its practical success, the stability of the method is only recently proved by utilizing the “spreading” property of the collision operator. In this talk, we introduce a new proof based on a careful $L^2$ estimate of the negative part of the solution. We also discuss the applicability of the result to various initial data, including both continuous and discontinuous functions. This talk is based on some joint works with Tong Yang (CityU) and Jingwei Hu (UW).

Add to calendar :

Date/Time : 22 Oct 2021 10:00
Venue : via Zoom
Speaker : Lee Hwee Kuan, Bioinformatics Institute, Agency for Science, Technology and Research

Join Zoom Meeting
https://nus-sg.zoom.us/j/9775606531?pwd=cVZ6eTRYS3ZzVUlFNXRuSlhMVjJSUT09

Meeting ID: 977 560 6531
Passcode: 123321

Abstract:  A novel weakly supervised tasks has been developed inspired by an application on cancer diagnosis. Given a bag of instances, we ask the question of how many unique classes of objects there are in the bag. This task was presented for the first time in our recent paper. Developing this task was inspired by the need to ask this question. “Given a cancer tissue sample, how many type of cancer cells there are in this sample”. We call this unique class count and formally proved that unique class count problem can be solved using a deep neural network in the multiple instance learning paradigm. We further apply our algorithm to prediction tumour purity in a cancer tissue sample. Tumour purity is the proportion of cancer cells in the tumour tissue. An accurate tumour purity estimation is crucial for accurate pathologic evaluation and for sample selection to minimize normal cell contamination in high throughput genomic analysis. Moreover, spatial tumour purity maps can help better understand the tumour microenvironment as a key determinant in tumour formation and therapeutic response.

Add to calendar :

Date/Time : 15 Oct 2021 10:00
Venue : via Zoom
Speaker : Ju Sun, University of Minnesota

Join Zoom Meeting
https://nus-sg.zoom.us/j/9775606531?pwd=cVZ6eTRYS3ZzVUlFNXRuSlhMVjJSUT09

Meeting ID: 977 560 6531
Passcode: 123321

Abstract:  Phase retrieval (PR) concerns the recovery of a signal (1D, 2D, or 3D) from its Fourier magnitudes, and is central to numerous areas of scientific imaging. Over the past decade, there have been numerous breakthroughs in understanding simplified or modified versions of the problem, but the existence of effective and robust methods for the original PR problem remains elusive and so theoretical insights are out of the question. In this talk, I will discuss why PR is difficult, the strength and limitation of conventional formulations, the promise and perils of deep learning approaches, as well as our recent methods that solve practical PR problems in an unprecedented regime.

Add to calendar :

Date/Time : 08 Oct 2021 10:00
Venue : via Zoom
Speaker : Peter Mueller, University of Texas at Austin

Join Zoom Meeting
https://nus-sg.zoom.us/j/9775606531?pwd=cVZ6eTRYS3ZzVUlFNXRuSlhMVjJSUT09

Meeting ID: 977 560 6531
Passcode: 123321

Abstract: Nonparametric Bayesian methods allow for principled model-based inference on unknown functions, distributions and structures without restriction to specific parametric models. The flexible prediction rules and the more realistic and complete description of related uncertainties are useful in many examples of clinical trial design and related data analysis. In this talk we review four related examples. The first example involves the comparison of two event time distributions for the time until resolution of air leaks after lung surgery. The second example is about inference for dynamic treatment regimes. Another example is a Bayesian basket trial design with a nonparametric Bayesian survival regression to find benefiting subpopulations. A last example is about the construction of a synthetic control arm to supplement a single-arm trial using real world data.

Add to calendar :

Date/Time : 01 Oct 2021 10:00
Venue : via Zoom
Speaker : Li Cheng, University of Alberta

Join Zoom Meeting
https://nus-sg.zoom.us/j/9775606531?pwd=cVZ6eTRYS3ZzVUlFNXRuSlhMVjJSUT09

Meeting ID: 977 560 6531
Passcode: 123321

Abstract:  Recent advancement of visual behavior analysis has opened door to many interesting applications. In this talk, I will discuss our research efforts toward addressing the related tasks of 3-D pose and shape estimation, action quality assessment, and motion synthesis of humans and beyond, from images and videos. Looking forward, our results could be applied to everyday life scenarios such as natural user interface, AR/VR, robotics, and gaming, among others.

Add to calendar :

Date/Time : 24 Sep 2021 10:00
Venue : via Zoom
Speaker : Weijie Su, University of Pennsylvania

Join Zoom Meeting
https://nus-sg.zoom.us/j/9775606531?pwd=cVZ6eTRYS3ZzVUlFNXRuSlhMVjJSUT09

Meeting ID: 977 560 6531
Passcode: 123321

Abstract:  The remarkable development of deep learning over the past decade relies heavily on sophisticated heuristics and tricks. To better exploit its potential in the coming decade, perhaps a rigorous framework for reasoning deep learning is needed, which however is not easy to build due to the intricate details of modern neural networks. For near-term purposes, a practical alternative is to develop a mathematically tractable surrogate model that yet maintains many characteristics of deep learning models.

This talk introduces a model of this kind as a tool toward understanding deep learning. The effectiveness of this model, which we term the Layer-Peeled Model, is evidenced by two use cases. First, we use this model to explain an empirical pattern of deep learning recently discovered by David Donoho and his students. Moreover, this model predicts a hitherto unknown phenomenon that we term Minority Collapse in deep learning training. This is based on joint work with Cong Fang, Hangfeng He, and Qi Long.

Add to calendar :