Researchers trained an AI model to ‘think’ like a baby, and it suddenly excelled.

 

Typically, AI models start with a blank slate and are trained on data with many different examples, from which the model constructs knowledge. But research on infants suggests this is not what babies do. Instead of building knowledge from scratch, infants start with some principled expectations about objects.

The exciting finding by Piloto and colleagues is that a deep-learning AI system modelled on what babies do, outperforms a system that begins with a blank slate and tries to learn based on experience alone.

To read the full article online, please visit The Conversation

News, Events and Highlights

Jun 2
Jul 12
Aug 10
Aug 19

BabyLab Podcasts

Starting from July 12, 2023 (weekly episodes).
Hosted by Jamila Rizvi, the eight-part series discusses ground-breaking research from Western Sydney University experts in psychology, linguistics, speech and language, and explores topics such as baby talk, bilingualism, learning to read, first words and play.