1min snip

Machine Learning Street Talk (MLST) cover image

#86 - Prof. YANN LECUN and Dr. RANDALL BALESTRIERO - SSL, Data Augmentation, Reward isn't enough [NEURIPS2022]

Machine Learning Street Talk (MLST)

NOTE

Creating the Adjacency Matrix

Creating an adjacency matrix in self-supervised learning reveals the representational capacity of embeddings. However, needing class labels to create the matrix raises questions. The surrogate task of distorting images in the self-supervised phase helps the final task if the eigenvectors of the two matrices overlap. The robustness of self-supervised representations is a current topic of interest.

00:00

Get the Snipd
podcast app

Unlock the knowledge in podcasts with the podcast player of the future.
App store bannerPlay store banner

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode

Save any
moment

Hear something you like? Tap your headphones to save it with AI-generated key takeaways

Share
& Export

Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode