From deep network mysteries to physics | The Royal Society
Ғылым және технология
The Royal Society Milner Award Lecture 2023 given by Professor Stéphane Mallat.
The remarkable performances of deep neural network remains a mathematical mystery. How come similar network architectures can capture properties of data as different as languages, natural images or physical fields? Physics provides a rich framework to understand this mystery. This presentation bridges neural network and physics, by showing in what sense they rely on similar mathematical principles. It introduces models of deep neural networks and complex physical fields, by separating phenomena appearing at different scales. Interactions of structures across scales are learned with random weights and wavelets. These models are applied to image classification, as well as generation of fluid turbulence and cosmological fields.
The Royal Society is a Fellowship of many of the world's most eminent scientists and is the oldest scientific academy in continuous existence.
▶royalsociety.org/
🔔Subscribe to our channel for exciting science videos and live events, many hosted by Brian Cox, our Professor for Public Engagement: bit.ly/3fQIFXB
We’re also on Twitter ▶ / royalsociety
Facebook ▶ / theroyalsociety
Instagram ▶ / theroyalsociety
And LinkedIn ▶ / the-royal-society
Пікірлер: 4
Great presentation
That was just extremely modest for me.
Q: Do AI's hallucinate? A: Yes, far too often to be considered reliable arbiters of reality.
Thank you for the upload interesting lecture and presentation. Imagine if humanity just focused more on how simplistic such issues are when simplifying such understandings merely requires toning down the intellectual lexicon of science language jargon and once again simply get to the fundamental route of the problems us humans usually tend to make incredibly difficult for most to understand and comprehend all while avoiding simple remedies to simple humanity problems that are complicated only because most intelligence scholars avoid real life topics by cloaking them in unequivocal double speak talk while always avoiding the simple but difficult to admit age old questions affecting populations never choosing to gave presentation to common sense natural intelligence solutions instead creating old but new loops of artificial intelligence which is generic fake conversations with little to no substance in real world reality problem solving but rather creating even more future complicated lies that kill science rather than reveal the con-sciences destroying the natural real world intelligence of tomorrow for more destructive reconstructive artificial intelligence of today. Either way still no truth when avoiding the simple truth to get to the facts of life. Just my thoughts. Life is simple as loving thy neighbor but we make it hard to sound smart to others all the time lying to our smart selves. No Thx Much ❤ & God Bless