To Join via Zoom: To join this seminar, please request Zoom connection details from headsec@stat.ubc.ca.

Abstract: The sparse group-lasso is an advanced statistical technique that involves the lasso and group-lasso penalties, which enables the sparsity at both the individual and group level regarding the natural grouping structure when solving high-dimensional learning problems. However, estimating the regularization paths can be computationally hard if the designed matrix input is large. To improve the computational efficiency, we develop an R package ’sparsegl' according to an algorithm by iteratively applying the KKT Stationarity Condition and the Strong Rule, which helps to identify the active and inactive predictors before updating their coefficients. Furthermore, we empirically illustrate its efficiency by comparing the running time to other existing packages for solving lasso-type problems.

Zoom Link & Talk Details

This talk is one of SSC Data Science & Analytics Section 2022 Webinar Series.

Zoom Link: https://uwaterloo.zoom.us/j/96812177576 (Passcode: DSA2022)

Talk Title: Piecing together the statistical puzzle to build a model

Abstract: As a student well into my statistics education, I knew a fair share of statistical methods and how they worked. But I couldn't see how they all fit together in a bigger framework. Methods like linear regression, machine learning, and out-of-sample scoring seemed to stand alone. In this talk, I'll discuss how your analysis' overarching statistical question, together with the nature of your data, tells us where and when to use what statistical methods. A variety of these methods will be discussed in the context of uncertainty quantification, tying statistics back to its probabilistic roots. By the end, you'll learn how to make clearer decisions when building a statistical model.

To join via Zoom: To join this seminar, please request Zoom connection details from headsec@stat.ubc.ca.

Presentation 1

Time: 4:00pm – 4:30pm

Speaker: Jeffrey Yiu, UBC Statistics MSc student

Title: Comparing public and private long-term care outcomes in British Columbia

Abstract: In British Columbia, approximately 90% of long-term care beds for seniors are publicly subsidized, while the remaining beds are privately paid. A 2018 report from the Office of the Seniors Advocate, an independent office of the B.C. provincial government, found that there were differences in emergency department visits, hospitalization rates, and mortality, depending on whether beds were publicly subsidized or privately paid. During my co-op terms with the B.C. Ministry of Health, I explored health databases such as the Discharge Abstract Database (DAD) and the National Ambulatory Care Reporting System (NACRS) to compare health outcomes between public and private long-term care clients in the 2019/20 fiscal year. Areas of investigation included demographics, morbidity, mortality, and utilization of health services.

Presentation 2

Time: 4:30pm – 5:00pm

Speaker: Pramoda Jayasinghe, UBC Statistics MSc student

Title: Working with health data and consulting for health sector clients

Abstract: Health data helps inform many decisions ranging from public health policies to showing the efficacy of a new drug. Due to the sensitivity of health data and the expectations of clients using said data, consulting for health-related projects has a unique set of challenges. While working on multiple projects during my co-op at Broadstreet, an organization that provides consulting services for health economics and outcomes research, I got to experience some of the practical issues that exist when consulting for health sector clients, specifically. The main objective of this presentation is to point out some aspects to consider when working with real-world health data.

Presentation 3

Time: 5:00pm – 5:30pm

Speaker: Li Zha, UBC Statistics MSc student

Title: Aspects of working with data between industry and health research

Abstract: Machine learning has been widely adopted in many applications in industry and research in recent years.  In this presentation, I will first introduce my industrial experience as a data science intern with TD Personal Banking where I mainly focused on building a propensity score matching for part of their test tool, and exploring data pruning to deal with noisy texts for fraud detection. Afterwards, I switched to the health industry by working on a polygenic risk score using GWAS (Genome-wide genetic association studies) data under Prof. Park’s lab in BC Cancer.  This presentation will focus more on sharing my learnings with working in different industries and provide some insights to those who are interested in data science roles.

To join via Zoom: To join this seminar, please request Zoom connection details from headsec@stat.ubc.ca.

Presentation 1

Time: 4:00pm – 4:30pm

Speaker: Shirley Cui, UBC Statistics MSc student

Title: Co-op experience in a Contract Research Organization(CRO) company: phase I-III clinical trials

Abstract: I started my industrial experience with a junior statistician position in a CRO company that provides clinical trial management services (eg. clinical research project management, on-site management, data management, statistical analysis, etc.) for the pharmaceutical industry companies. I mainly assist senior statistician with study design, sample size calculation, randomization list generation, eCRF review and annotation etc. I would like to share my coop experience in participating research projects involving phase I-III clinical trials and provide insights to those who are interested in pursuing a career in this field.

Presentation 2

Time: 4:30pm – 5:00pm

Speaker: Shuyi Tan, UBC Statistics MSc student

Title: Data Analytics in Healthcare

Abstract: The healthcare industry collects vast amounts of health data. Analysis of these data informs important decisions related to healthcare delivery, evaluation, and quality management. For the past 12 months, Shuyi has been working as a business analyst at Fraser Health Authority focusing on process improvement and data analysis. She would like to talk about her experience working with healthcare data in terms of data acquisition, data mining, insight extraction, and potential challenges. Since the data-related inquiries from clinical specialists will be quite different from those from an academic setting, examples would be provided to give some insights into how to approach these real-world questions.

Presentation 3

Time: 5:00pm – 5:30pm

Speaker: Zhipeng Zhu, UBC Statistics MSc student

Title: Practical Tools for Junior Data Analyst: From Graphic Software to DSLs

Abstract: This presentation talks about some experience as an entry-level data analyst or data scientist and introduces the idea of participating in DSL development.

Compared to academic tasks in many statistics courses, the actual work content as an analyst or consultant is quite different. While statistic knowledge is indispensable, using tools flexibly has become more critical in many workloads. To deal with complicated cases, analytic software with graphic UI has become an essential tool for most analysts. These graphic software offer excellent plot outputs and a superior feature of document generation compared to coding style tools such as R. However, graphic software sometimes cost the performance for particular tasks, and hence we require the hard-coded programs. This inspires me to participate in developing R packages “distplyr” and “distionary” as DSLs with Vincenzo Coia. Even though we still need to use coding languages for many specific tasks, a good DSL will significantly improve our productivity.

To join via Zoom: To join this seminar, please request Zoom connection details from headsec@stat.ubc.ca

Title: Metropolis Adjusted Underdamped Langevin Trajectories: a robust alternative to Hamiltonian Monte-Carlo

Abstract: Sampling approximations for high dimensional statistical models often rely on so-called gradient-based MCMC algorithms. It is now well established that these samplers scale better with the dimension than other state of the art MCMC samplers, but are also more sensitive to tuning [5]. Among these, Hamiltonian Monte Carlo is a widely used sampling method shown to achieve gold standard d^{1/4} scaling with respect to the dimension [1]. However it is also known that its efficiency is quite sensible to the choice of integration time, see e.g. [4], [2]. This problem is related to periodicity in the autocorrelations induced by the deterministic trajectories of Hamiltonian dynamics. To tackle this issue, we develop a robust alternative to HMC built upon underdamped Langevin (namely Metropolis Adjusted Underdamped Langevin Trajectories, or MAULT), inducing randomness in the trajectories through a continuous refreshment of the velocities. We study the optimal scaling problem for MAULT and recover the d^{1/4} scaling of HMC proven in [1] without additional assumptions. Furthermore we highlight the fact that autocorrelations for MAULT can be controlled by a uniform and monotonous bound thanks to the randomness induced in the trajectories, and therefore achieves robustness to tuning. Finally, we compare our approach to Randomized HMC ([2], [3]) and establish quantitative contraction rates for the 2-Wasserstein distance that support the choice of underdamped Langevin dynamics.

This is a joint work with Jure Vogrinc (University of Warwick)

To join via Zoom: To join this seminar, please request Zoom connection details from headsec@stat.ubc.ca.

Presentation 1

Time: 11:00am – 11:30am

Speaker: Grace Yin, UBC Statistics MSc student

Title: On the Relationship between Predictive Models and Structural Causal Model Learning

Abstract: Structural Causal Models (SCMs) are a key component in causal inference and they have been used for a long time in many fields. We proposed an approach to identify the structure of an SCM based on a constant risk theorem. In particular, we proved that for an equivariant model and predictor, the risk function is constant across interventions described by the action of a group. These theories give a straightforward understanding of certain types of causal model identification. We also explored the risks on a specific SCM for linear regression predictive models with different types of interventions through simulation experiments.

Presentation 2

Time: 11:30am – 12:00pm

Speaker: Wei Tang, UBC Statistics MSc student

Title: Improved model training for multi-horizon time series forecasting in the context of COVID-19

Abstract: Predicting how the Covid-19 pandemic evolves in the future is very important for public health workers and policymakers to prepare for it. The prediction task is a multi-step time series forecasting task. In this report, we compared two commonly used strategies, the iterative and direct strategy, for this task on the Covid dataset under various experiment settings. We further enhanced the two strategies with ideas from K-Nearest Neighborhood (KNN) algorithm to construct the training dataset and proposed an improved iterative strategy which we call the dynamic iterative strategy. The proposed KNN enhanced forecasting strategies are significantly better than the native iterative and direct strategies and achieve satisfying prediction accuracy on the covid prediction challenge at the Covid-19 Forecasting Hub.

To join via Zoom: To join this seminar, please request Zoom connection details from headsec@stat.ubc.ca

Title: Bayesian Dose-Finding Design for Molecularly Targeted Agents and Immunotherapy

Abstract: Molecularly targeted agents and immunotherapy have revolutionized modern cancer treatment. Unlike chemotherapy, the maximum tolerated dose of the targeted therapies may not pose significant clinical benefit over the lower doses. By simultaneously considering both binary toxicity and efficacy endpoints, phase I/II trials can identify a better dose for subsequent phase II trials than traditional phase I trials in terms of efficacy-toxicity tradeoff. Existing phase I/II dose-finding methods are model-based or need to pre-specify many design parameters, which makes them difficult to implement in practice. To strengthen and simplify the current practice of phase I/II trials, we propose a utility-based toxicity probability interval (uTPI) design for finding the optimal biological dose (OBD) where binary toxicity and efficacy endpoints are observed. The uTPI design is model-assisted in nature, simply modeling the utility outcomes observed at the current dose level based on a quasibinomial likelihood. Toxicity probability intervals are used to screen out overly toxic dose levels, and then the dose escalation/de-escalation decisions are made adaptively by comparing the posterior utility distributions of the adjacent levels of the current dose. The uTPI design is flexible in accommodating various utility functions while only needing minimum design parameters. A prominent feature of the uTPI design is that it has a simple decision structure such that a concise dose-assignment decision table can be calculated before the start of the trial and be used throughout the trial, which greatly simplifies the practical implementation of the design. Extensive simulation studies demonstrate that the proposed uTPI design yields desirable as well as robust performance under various scenarios.

This talk is based on the joint work with Ruitao Lin and Ying Yuan at MD Anderson Cancer Center.

To join via Zoom: Please request Zoom connection details from headsec@stat.ubc.ca

Title: Stochastic Geometry for Machine Learning

Abstract: The Mondrian process in machine learning is a recursive partition of space with random axis-aligned cuts used to build random forests and Laplace kernel approximations. The construction allows for efficient online algorithms, but the restriction to axis-aligned cuts does not capture dependencies between features. By viewing the Mondrian as a special case of the stable under iterated (STIT) process in stochastic geometry, we resolve open questions about the generalization of cut directions. We utilize the theory of stationary random tessellations to show that STIT random features approximate a large class of stationary kernels and STIT random forests achieve minimax rates for Lipschitz and C^2 functions. This work opens many new questions at the intersection of stochastic geometry and machine learning. Based on joint work with Ngoc Tran.

Registration & talk details

This talk is one of Data Science Applied Research and Education Seminar (ARES) series. Learn more and register for this talk here.

Talk Title: Understanding animal movement with state-space models

Abstract: Movement data have become essential for our understanding of animal ecology. However, such data are associated with a broad range of challenges. For example, tracking the movement of many species (e.g. fish and small birds) is still limited to inaccurate technology, such as light-based geolocation. Making behavioral and spatial inferences based on such data is difficult because the tracks they create are not good representations of the animals’ movement. Understanding the behavior of animals using movement data is challenging even with accurate movement data, because we are often making inference on a process that is not observed directly. Through a set of examples from a broad range of animals (fish, bears, marine mammals), I will demonstrate how a state-space modeling framework can improve our understanding of animal movement.

To Join Via Zoom: To join this seminar, please request Zoom connection details from headsec@stat.ubc.ca.

Abstract: MCMC methods are a popular tool in computation science used to evaluate expectations with respect to complex probability distributions over general state spaces. They work by averaging over the trajectory of a Markov chain stationary with respect to the target distribution. In theory, the MCMC algorithms converge asymptotically, but, in practice, for challenging problems where the target distributions are high-dimensional with well-separated modes, MCMC algorithms can get trapped exploring local regions of high probability and suffer from poor mixing.

Physicists and statisticians independently introduced parallel tempering (PT) algorithms to tackle this issue. PT delegates the task of exploration to additional annealed chains running in parallel with better mixing properties. They then communicate with the target chain of interest and help discover new unexplored regions of the sample space. Since their introduction in the 90s, PT algorithms are still extensively used to improve mixing in challenging sampling problems arising in statistics, physics, computational chemistry, phylogenetics, and machine learning.

The classical approach to designing PT algorithms was developed using a reversible paradigm that is difficult to tune and deteriorates in performance when too many parallel chains are introduced. This talk will introduce a new non-reversible paradigm for PT that dominates its reversible counterpart while avoiding the performance collapse endemic to reversible methods. We will then establish near-optimal tuning guidelines and efficient black-box methodology scalable to GPUs. Our work out-performs state-of-the-art PT methods and has been used at scale by researchers to study the evolutionary structure of cancer and discover magnetic polarization in the photograph of the supermassive black hole M87.