Developmental Interpretability, stemming from Singular Learning Theory (SLT), aims at explaining behaviours of AI models through a generalised framework. If SLT is applicable for our current models, that would allow us to finally tackle the largest problems we face such as deception of AI models. I research on testing SLT’s claims on gradient descents, particularly that Stochastic gradient descent general using better than natural gradient descent. This test is necessary in… Show more Developmental Interpretability, stemming from Singular Learning Theory (SLT), aims at explaining behaviours of AI models through a generalised framework. If SLT is applicable for our current models, that would allow us to finally tackle the largest problems we face such as deception of AI models. I research on testing SLT’s claims on gradient descents, particularly that Stochastic gradient descent general using better than natural gradient descent. This test is necessary in building evidence for a possible “grand unified theory of AI” Show less

Educate fellows of AI Safety Technical Fellowship on latest technical AI research papers This fellowship is a 7 week course where I conduct lessons for fellows ranging from undergraduates to PHDs

I co-started Hydraited. We are developing Water Quality checks that can be used by rural communities, empowering them to monitor their own water quality. I alone developed the technological system used to run our product, powered by my custom AI model

Built a “light weight” Regional-CNN Model from scratch to detect and classify arrows in a financial chart. Used further statistical methods (k-means clustering, text recognition) to provide more information on the chart

I successfully completed command school and served a sergeant in the Singapore Army