Toggle light / dark theme

According to a statement by the firm, the study showed that AI models trained on hand images achieve comparable accuracy to those using facial images, with an average error of 4.1 and 4.7 years in predicting chronological age.

The AI model in the study was primarily trained by employing the Indian population dataset to ensure representation of diverse skin tones and address AI’s bias challenges, especially pertaining to ethnicity-specific considerations in age prediction.

By focusing on the Indian population, the study aimed to develop an AI model tailored to this demographic, mitigating biases and promoting fairer AI solutions. Additionally, the research’s market relevance in India’s growing skincare and AI sectors underscores the strategic importance of using an Indian dataset for this study.

Chapters 00:00 — Intro + Background 05:06 — From KART to KAN 07:56 — MLP vs KAN 16:05 — Accuracy: Scaling of KANs 26:35 — Interpretability: KAN for Science 38:04 — Q+A Break 57:15 — Strengths and Weaknesses 59:28 — Philosophy 1:08:45 — Anecdotes Behind the Scenes…


Portal is the home of the AI for drug discovery community. Join for more details on this talk and to connect with the speakers: https://portal.valencelabs.com/logg.

Abstract: Inspired by the Kolmogorov-Arnold representation theorem, we propose Kolmogorov-Arnold Networks (KANs) as promising alternatives to Multi-Layer Perceptrons (MLPs). While MLPs have fixed activation functions on nodes (\.