Up the Ante, Down the Dimensionality: Masterclass 2026 by Assoc Prof Duane
by Ariel Joshua Lau, Warren Wee, Yeo Chi Enn Talia, Danyson Wong De Sheng
With thunderous applause, the SIMC2.0 2026 Masterclass commenced at 10.20 a.m. on 25th May. Conducted by Associate Professor and Chief Setter Duane Loh (Ne-Te), the masterclass was the first major event for participants on the Endeavour track following the Opening Ceremony – where they dove deep into the mathematical aspects of datasets in AI computation.
The masterclass began with multidimensional objects and the unique properties that they exhibit at high dimensions and swiftly proceeded into exploring covariance in the analysis of large datasets with the power method. Following demonstrations with spectroscopy diagrams (and a tennis racket!), the lecture eventually culminated in Principal Component Analysis (PCA).
Participants mentioned that the masterclass was complicated to follow as it covered university and even PhD-level topics at some points. Despite that, they were fascinated by the depth of the content!
“These competitions are a nice place to bring in these more sophisticated, bigger ideas as it is supposed to be at a more advanced level,” explained Prof Duane in an exclusive interview with Epigraph. The professor shared his design philosophy behind crafting the Masterclass: providing the concrete component to understand matrices and linear algebra in a spatial context, as well as showing off the power of using matrices.
He left participants with two pieces of advice: to learn to be “bilingual” in AI and Science, and to utilise AI as a powerful trainer.
“Don’t be like him there,” Prof Duane mentioned in his lecture, gesturing to a man losing his bearings in a murky haze of equations. “Move inside the black box as much as possible.”
The masterclass concluded with the professor wishing participants good luck for the challenges awaiting them, as well as encouraging them to embrace diversity.
“PCA assumes the manifold is flat. But the real adventure begins when it isn’t.”
