NB! HDR talks take place on Tuesdays at 10am in room HN 3.41.

Please note that AI/ML cluster holds its regular seminars here:


Please note that Foundations cluster holds its regular seminars here:


We expect Data Science and Computational cluster oral presentations to take place in this common time slot.

Please email HDR Admin to book a slot for your Oral Presentation milestone.

Upcoming Talks

Past Talks

5-6pm, 30 May 2024 (Thursday)

HN 3.41 in the Hanna Neumann Building

PhD Oral Presentation: Zejun Zhang

Title: Comprehensive Analysis of Programming Language Idioms and Their Application to Software Quality

Abstract: Achieving high-quality software is fundamental to the success and sustainability of software development projects. Software quality encompasses multiple facets, including correctness, maintainability, readability and performance, each posing its unique challenges. Achieving these goals requires a deep understanding of best practices and effective coding conventions. Programming idioms, celebrated for their conciseness and improved performance, offer substantial benefits but also introduce complexity that can hinder their adoption. This thesis explores the enhancement of software quality through the strategic use of Pythonic idioms, focusing on three critical aspects: code maintainability, readability, and performance optimization.

We first introduce RIdiom, an automated refactoring tool, to transform non-idiomatic Python code into idiomatic forms, enhancing maintainability and reducing manual effort. By combining rule-based methods with LLM adaptability, we boost code detection and refactoring accuracy, promoting consistent best practices across codebases. Additionally, we provide explanations of Pythonic idioms in non-idiomatic code, aiding comprehension, especially for developers less familiar with Pythonic idioms. Furthermore, our empirical study evaluates the performance impacts of Pythonic idioms, offering evidence-based guidelines for optimal code performance. Our findings provide nuanced insights into the benefits and limitations of Pythonic idioms, supporting developers in writing more correct, maintainable, and efficient code.

Bio: Zejun Zhang is a final year PhD student at the ANU’s School of Computing. She is dedicated to pushing the boundaries of high-quality software through a multidisciplinary approach. She has authored 10 papers in top-ranked international conferences and journals. Five of them are the first author and published in Core A* venues such as ICSE and FSE. Her contributions have been recognized with an ACM SIGSOFT Distinguished Paper Award at ICSE 2024.

10-11am, 21 May 2024 (Tuesday)

N3.41, Hanna Neumann Building 145

PhD Oral Presentation: Jonathan Ting

Title: Automated Sampling and Labelling of High-throughput Nanomaterials Data for Machine Learning and Inference

Abstract: With the accumulation of larger data sets, a lot of scientific disciplines have arrived at a point where the fourth paradigm of scientific discovery is viable. However, the bottleneck for the extraction of actionable insights from scientific data often lies in the sampling, labelling, and causal inference of the data. In this talk, I will cover methodologies that I have developed over my Ph.D. using nanomaterials data as testbed, for (i) the sampling of comprehensive data sets with minimal selection bias and instance redundancy, (ii) the automatic labelling of nanomaterials data with affordable and relevant labels including box-counting dimensions and unsupervised learnt surface patterns, and (iii) the extraction of causal paths leading to properties of interest from the data.

Bio: Jonathan Ting is a final-year Ph.D. student at the School of Computing, ANU College of Engineering, Computing & Cybernetics, under the supervision of Prof. Amanda Barnard, Prof. Sean Smith, Prof. Nick Birbilis, and Prof. Andrew Wood. He received his Bachelor of Advanced Science degree with first-class Honours and University Medal from the University of Queensland in 2019. His research interests include nanoinformatics, catalysis, molecular dynamics, high-performance computing, and machine learning.

12-1pm, 9 May 2024 (Thursday)

N3.41, Hanna Neumann Building 145

PhD Oral Presentation: Haolei Ye

Title: Hardware-Software Integrated Acceleration for Scientific Computing Applications

Abstract: As semiconductor manufacturing advances, hardware performance enhancement has shifted from improving single pipeline performance to exploiting parallelization. However, due to limited progress in developing automated parallel translators, programmers often have to rely on their intuitions to accelerate software performance by observing bottlenecks. Consequently, scientific applications may fail to fully utilize existing hardware platforms, leading to suboptimal performance. In this PhD oral presentation, I will propose three principles, namely PC2 principles, which provide theoretical analysis for hardware-software integrated acceleration on single-machine systems. These principles are verified by four real-world scientific software acceleration cases on different hardware platforms: 1) Drug Combination Pathway Network, 2) Polyploid DNA Scaffolding, 3) Structural Optical Flow, and 4) Real-time Machine Learning based Visual Object Tracking.

bars search times