Webinar Replay
Stress-testing Algorithms, Solvers, and Models via Instance Space Analysis
This talk will provide a brief introduction to Instance Space Analysis as a methodology that provides a mathematically rigorous foundation for “stress-testing” algorithms.
April 30, 2026
10:00 AM ET | 4PM CET
Webinar Replay
Stress-testing Algorithms, Solvers, and Models via Instance Space Analysis
This talk will provide a brief introduction to Instance Space Analysis as a methodology that provides a mathematically rigorous foundation for “stress-testing” algorithms.
April 30, 2026
10:00 AM ET | 4PM CET
Webinar Replay
Stress-testing Algorithms, Solvers, and Models via Instance Space Analysis
This talk will provide a brief introduction to Instance Space Analysis as a methodology that provides a mathematically rigorous foundation for “stress-testing” algorithms.
April 30, 2026
10:00 AM ET | 4PM CET
Event Recap
Missed the live session? Watch the replay to explore why building trust in optimisation algorithms has never been more important. From heuristics and exact solvers to model design and parameter tuning, modern decision-making relies heavily on algorithmic choices—but how can we be confident they perform reliably in practice?
This session introduces Instance Space Analysis (ISA), a rigorous methodology for stress-testing algorithms. ISA helps uncover where algorithms perform well—and where they don’t—by evaluating them across a wide range of unbiased, diverse problem instances.
You’ll also see how ISA is being used in practice through an accessible online tool, adopted by researchers worldwide and industry teams aiming to reduce risk when deploying critical algorithms.
Through real-world case studies in areas like timetabling and bin packing, the webinar demonstrates how ISA can generate meaningful insights into the performance of heuristics, MIP models, and solvers—helping you better understand algorithm behaviour under different conditions.
Event Recap
Missed the live session? Watch the replay to explore why building trust in optimisation algorithms has never been more important. From heuristics and exact solvers to model design and parameter tuning, modern decision-making relies heavily on algorithmic choices—but how can we be confident they perform reliably in practice?
This session introduces Instance Space Analysis (ISA), a rigorous methodology for stress-testing algorithms. ISA helps uncover where algorithms perform well—and where they don’t—by evaluating them across a wide range of unbiased, diverse problem instances.
You’ll also see how ISA is being used in practice through an accessible online tool, adopted by researchers worldwide and industry teams aiming to reduce risk when deploying critical algorithms.
Through real-world case studies in areas like timetabling and bin packing, the webinar demonstrates how ISA can generate meaningful insights into the performance of heuristics, MIP models, and solvers—helping you better understand algorithm behaviour under different conditions.
Event Recap
Missed the live session? Watch the replay to explore why building trust in optimisation algorithms has never been more important. From heuristics and exact solvers to model design and parameter tuning, modern decision-making relies heavily on algorithmic choices—but how can we be confident they perform reliably in practice?
This session introduces Instance Space Analysis (ISA), a rigorous methodology for stress-testing algorithms. ISA helps uncover where algorithms perform well—and where they don’t—by evaluating them across a wide range of unbiased, diverse problem instances.
You’ll also see how ISA is being used in practice through an accessible online tool, adopted by researchers worldwide and industry teams aiming to reduce risk when deploying critical algorithms.
Through real-world case studies in areas like timetabling and bin packing, the webinar demonstrates how ISA can generate meaningful insights into the performance of heuristics, MIP models, and solvers—helping you better understand algorithm behaviour under different conditions.
Speaker
Meet Your Expert Speaker
Learn from the best in the industry.
Kate Smith-Miles
Pro Vice-Chancellor (Research Capability)
Professor Kate Smith-Miles is a leading Australian mathematician and the Pro Vice-Chancellor (Research Capability) at The University of Melbourne, where she also holds a Melbourne Laureate Professorship in Applied Mathematics. She directs the ARC Training Centre OPTIMA, which focuses on optimisation technologies.
With a PhD in Electrical Engineering and a B.Sc(Hons) in Mathematics from The University of Melbourne, her academic career began in 1996. She has authored two books and around 300 publications in neural networks, optimisation, and data mining, and has supervised over 30 PhD students. Her leadership roles have included positions at Monash University, President of the Australian Mathematical Society, and ARC College of Experts member.
She has secured over AUD$20 million in grants, including a prestigious Georgina Sweet Laureate Fellowship. Her Instance Space Analysis work led to the development of MATILDA, a benchmarking tool for algorithms. Professor Smith-Miles has received numerous accolades, including the AO in the 2024 King’s Birthday Honours and fellowships in major scientific societies.
A dedicated mentor and advocate for women in STEM, she also consults widely with industry, applying her expertise in optimisation and intelligent systems to real-world challenges. Her career reflects excellence in research, leadership, and service to the mathematical sciences.
Speaker
Meet Your Expert Speaker
Learn from the best in the industry.
Kate Smith-Miles
Pro Vice-Chancellor (Research Capability)
Professor Kate Smith-Miles is a leading Australian mathematician and the Pro Vice-Chancellor (Research Capability) at The University of Melbourne, where she also holds a Melbourne Laureate Professorship in Applied Mathematics. She directs the ARC Training Centre OPTIMA, which focuses on optimisation technologies.
With a PhD in Electrical Engineering and a B.Sc(Hons) in Mathematics from The University of Melbourne, her academic career began in 1996. She has authored two books and around 300 publications in neural networks, optimisation, and data mining, and has supervised over 30 PhD students. Her leadership roles have included positions at Monash University, President of the Australian Mathematical Society, and ARC College of Experts member.
She has secured over AUD$20 million in grants, including a prestigious Georgina Sweet Laureate Fellowship. Her Instance Space Analysis work led to the development of MATILDA, a benchmarking tool for algorithms. Professor Smith-Miles has received numerous accolades, including the AO in the 2024 King’s Birthday Honours and fellowships in major scientific societies.
A dedicated mentor and advocate for women in STEM, she also consults widely with industry, applying her expertise in optimisation and intelligent systems to real-world challenges. Her career reflects excellence in research, leadership, and service to the mathematical sciences.
Speaker
Meet Your Expert Speaker
Learn from the best in the industry.
Pro Vice-Chancellor (Research Capability)
Kate Smith-Miles
Professor Kate Smith-Miles is a leading Australian mathematician and the Pro Vice-Chancellor (Research Capability) at The University of Melbourne, where she also holds a Melbourne Laureate Professorship in Applied Mathematics. She directs the ARC Training Centre OPTIMA, which focuses on optimisation technologies.
With a PhD in Electrical Engineering and a B.Sc(Hons) in Mathematics from The University of Melbourne, her academic career began in 1996. She has authored two books and around 300 publications in neural networks, optimisation, and data mining, and has supervised over 30 PhD students. Her leadership roles have included positions at Monash University, President of the Australian Mathematical Society, and ARC College of Experts member.
She has secured over AUD$20 million in grants, including a prestigious Georgina Sweet Laureate Fellowship. Her Instance Space Analysis work led to the development of MATILDA, a benchmarking tool for algorithms. Professor Smith-Miles has received numerous accolades, including the AO in the 2024 King’s Birthday Honours and fellowships in major scientific societies.
A dedicated mentor and advocate for women in STEM, she also consults widely with industry, applying her expertise in optimisation and intelligent systems to real-world challenges. Her career reflects excellence in research, leadership, and service to the mathematical sciences.
