Datum / čas
Date(s) - 01.10.
14:00 - 15:30

Kategorie ne Kategorie

Dear colleagues,

I am happy to announce that on Wednesday, October 1, at 14.00 will be given a talk by

Prof. Vladik Kreinovich,

Univ. of Texas at El Paso

Spectacular Successes of Deep Learning: Why and What Next?

Abstract: Successes of deep learning are partly due to appropriate selection of activation function, pooling functions, etc. Most of these choices have been made based on empirical comparison and heuristic ideas. In the first – why — part of this talk, we show that many of these choices – and the surprising success of deep learning in the first place – can be explained by reasonably simple and natural mathematics.

In the second – what next — part, we present ideas on how to overcome limitations of current deep learning techniques. One of these limitations is that, in contrast to humans who can learn from a few examples and learn fast, modern deep learning techniques require a large amount of data to learn, and they take a long time to train. In this talk, we show that neural networks do have a potential to learn from a small number of examples – and learn fast. We speculate that the corresponding idea may already be implicitly implemented in Large Language Models – which may partially explain their (somewhat mysterious) success.

CIIRC – ROOM No. 101(near the reception in the building A).

The speaker will be introduced by Prof. Václav Hlaváč.

Best regards,

Vladimír Mařík