Upcoming Seminars

Speaker

Dr. Dieter Büchler, Amii Fellow & Assistant Professor in the Department of Computing Science at the University of Alberta

Title

The Role of the Robotic Body in Skills Learning

Abstract
Learning skills on physical robots to solve tasks at human-level performance is a key enabling technology for the future. Much of today’s research focuses on developing learning algorithms, while designing the robotic body used for execution is often treated as a separate problem. Biological systems, and especially the human anatomy, show that generalization and high performance in difficult tasks are facilitated by the body itself—an idea captured by the fields of embodied intelligence and morphological computation. In this talk, I will outline the key challenges in developing new learning approaches that leverage the intelligence inherent to the robotic body.

Presenter Bio
Dieter Büchler is an Assistant Professor in the Computing Science department at the University of Alberta and also leads a research group in the Empirical Inference department at the Max-Planck Institute for Intelligent Systems in Tübingen, Germany. Dieter holds a Canada CIFAR AI chair and is an Alberta Machine Intelligence (Amii) fellow. He earned a Ph.D. in computer science from the TU Darmstadt under Jan Peters and Bernhard Schölkopf and performed the research at the MPI for Intelligent Systems. While pursuing his PhD, Dieter interned at X, the Moonshot Factory (formerly Google X). He holds an M.Sc. in Biomedical Engineering from Imperial College London and a B.E. in Information and Electrical Engineering from HAW Hamburg with generous support from Siemens. His mission is to achieve human performance in athletic, rapidly changing, uncertain, and high-dimensional tasks with physical robots. His research group develops learning approaches for complex systems, like soft and muscular robots, which can excel in these demanding domains. The group also studies how the robotic body influences the acquisition of robotic skills.

Timing & Location PIZZA ROOM CHANGE

Pizza in UComm Room 2-480 from 11:30 to noon

Seminar in UComm Alfred Sorenson Community Hall Room 2-350 from noon to 1

Add event to calendar

Add event

Speaker

Dr. Christian Marius Lillelund, Postdoc in the Department of Molecular Medicine (MOMA) at Aarhus University, Denmark, hosted by Dr. Russ Greiner

Title

Learning Multi-Event Survival Models with Applications to ALS

Abstract
Many clinical prediction problems involve multiple non-exclusive events, yet most survival methods handle only single-event or competing-risks settings. We present MENSA, a multi-event survival model that jointly learns flexible time-to-event distributions while capturing dependencies and temporal ordering across events. We apply this framework to amyotrophic lateral sclerosis (ALS), modelling the progression of functional decline among patients. Here, MENSA generates individualized survival curves for loss of speaking, swallowing, handwriting, walking, and breathing ability, outperforming standard baselines and reflecting known clinical heterogeneity such as earlier bulbar decline in bulbar-onset patients. The model also enables counterfactual analyses to assess how covariate changes influence predicted outcomes. Overall, the results show that multi-event models can give more practical, fine-grained predictions that support better care decisions in ALS and other conditions with varied progression.

Presenter Bio
Many clinical prediction problems involve multiple non-exclusive events, yet most survival methods handle only single-event or competing-risks settings. We present MENSA, a multi-event survival model that jointly learns flexible time-to-event distributions while capturing dependencies and temporal ordering across events. We apply this framework to amyotrophic lateral sclerosis (ALS), modelling the progression of functional decline among patients. Here, MENSA generates individualized survival curves for loss of speaking, swallowing, handwriting, walking, and breathing ability, outperforming standard baselines and reflecting known clinical heterogeneity such as earlier bulbar decline in bulbar-onset patients. The model also enables counterfactual analyses to assess how covariate changes influence predicted outcomes. Overall, the results show that multi-event models can give more practical, fine-grained predictions that support better care decisions in ALS and other conditions with varied progression.

Timing & Location

UComm Seminar Room 2-108, pizza from 11:30, seminar from noon to 1

Add event to calendar

Add event

Speaker

Dr. Zhou Yang, Amii Fellow & Assistant Professor in the Department of Computing Science at the University of Alberta

Title

TBA

Abstract
TBA

Presenter Bio
TBA

Timing & Location

Pizza in UComm Seminar Room 2-108 from 11:30 to noon

Seminar in UComm Alfred Sorenson Community Hall Room 2-350 from noon to 1

Add event to calendar

TBA

Speaker

Dr. Irina Rish, Professor at the Université de Montréal, hosted by Dr. Rich Sutton

Title

TBA

Abstract
TBA

Presenter Bio
TBA

Website

TBA

Timing & Location ONE HOUR EARLIER

UComm Seminar Room 2-108, 11:00 am to noon

Add event to calendar

TBA

Speaker

J. Fernando Hernandez-Garcia, PhD student at the UofA, supervised by Dr. Rich Sutton

Title

TBA

Abstract
TBA

Presenter Bio
TBA

Website

TBA

Timing & Location

UComm Seminar Room 2-108, pizza from 11:30, seminar from noon to 1

Add event to calendar

TBA

NO SEMINARS - Winter Closure

Speaker

Dr. Tony Yousefnezhad, Senior Data Scientist at National Bank of Canada, hosted by Dr. Russ Greiner

Title

Orthogonal Contrastive Learning for Multi-Representation fMRI Analysis

Abstract
Functional MRI offers a powerful window into human cognition, yet challenges such as low signal-to-noise ratio, high dimensionality, and limited sample sizes remain major barriers—especially when integrating data across subjects or imaging sites. In this talk, we will introduce Orthogonal Contrastive Learning (OCL), a unified framework for aligning and analyzing multi-subject fMRI data without requiring temporal synchronization or equal time-series lengths.

OCL leverages two identical encoder networks: an online network trained with a contrastive objective that brings same-stimulus responses closer while separating different ones, and a target network that tracks the online model through an exponential moving average for stable learning. Each layer integrates QR decomposition for orthogonal feature extraction, locality-sensitive hashing (LSH) for compact subject-specific signatures, positional encoding for temporal-spatial fusion, and a transformer encoder for generating discriminative neural embeddings. I will also discuss OCL’s unsupervised pretraining on synthetic fMRI-like data and its transfer learning workflow for multi-site applications.

Presenter Bio
Dr. Tony Yousefnezhad is a Senior Data Scientist in the Department of Information Management at the National Bank of Canada, with cross-continental experience spanning Eurasia, East Asia, and North America. In addition to his industry role, he actively contributes to academic research and open-source innovation through his self-founded company, Learning By Machine. His research is at the forefront of pioneering advancements in machine learning, with a focus on deep learning, natural language processing (NLP), and reinforcement learning (RL) methodologies. These developments are designed to analyze a wide range of data modalities, including time series, text, images, audio, and wearable signals.

Website

Tony Muhammad Yousefnezhad

Timing & Location

UComm Seminar Room 2-108, pizza from 11:30, seminar from noon to 1

Add event to calendar

Add to calendar

Speaker

Shuai Liu, PhD student, University of Alberta, supervised by Dr. Csaba Szepesvári & Dr. Xiaoqi Tan

Title

Sample Complexity for Zero-Discounted MDP with Linear / Logistic Function Approximation

Abstract
We will discuss ideas behind algorithms that achieve nontrivial sample complexity guarantees for 0-discounted MDPs with linear/ logistic function approximation, a.k.a. stochastic contextual linear / logistic contextual bandits, including a deterministic UCB-like algorithm and a computationally efficient Thompson Sampling variant.

Presenter Bio
Shuai Liu is a PhD student in the Computing Science department at University of Alberta, co-supervised by Dr. Csaba Szepesvári and Dr. Xiaoqi Tan. His current research interest lies in reinforcement learning theory (policy gradient methods), bandit algorithms and optimization. Before that, he obtained his MSc in Computing Science at University of Alberta under the supervision of Dr. Szepesvári and obtained a Bachelor's degree in Computer Science at the Harbin Institute of Technology.

Website

Shuai Liu

Timing & Location

UComm Seminar Room 2-108, pizza from 11:30, seminar from noon to 1

Add event to calendar

Add event

Speaker

TBA

Speaker

Dr. Carrie Demmans Epp, Amii Fellow & Associate Professor in the Department of Computing Science at the University of Alberta

Title

TBA

Abstract
TBA

Presenter Bio
TBA

Timing & Location

Pizza in UComm Seminar Room 2-108 from 11:30 to noon

Seminar in UComm Alfred Sorenson Community Hall Room 2-350 from noon to 1

Add event to calendar

TBA