MIT 6.S191 (2018): Deep Learning Limitations and New Frontiers

preview_player
Показать описание
MIT Introduction to Deep Learning 6.S191: Lecture 6
Deep Learning Limitations and New Frontiers
Lecturer: Alexander Amini
January 2018

Рекомендации по теме
Комментарии
Автор

Amazing lecture, it helps not to be lost in current hype of deep learning and shows good DL research directions. I'm trying to research on autoML, but the problem is, it needs good computational power (GPU accelerated), which make it an expensive research area for students who don't have access for such resources. Big companies like google (deep mind) are doing a great job on AutoML. Thanks for sharing the lecture

abelmuna
Автор

Excellent! A perfect mix of teaching and lecturing.

ogsconnect
Автор

I think machine learning is good at solve some difficult tasks(but can get data), when machine learning itself become a difficult task, machine learning for machine learning. When my first time try to write a CNN, I really lose direction how to tune architecture.Great lecture!

张开焰
Автор

Thank you. So informative. However its a crash course :) Need to take time to absorb

peacebeuponhimever