Ian Goodfellow: Adversarial Machine Learning (ICLR 2019 invited talk)

Speaker: Ian Goodfellow
Abstract:
Until about 2013, most researchers studying machine learning for artificial intelligence all worked on a common goal: get machine learning to work for AI-scale tasks. Now that supervised learning works, there is a Cambrian explosion of new research directions: making machine learning secure, making machine learning private, getting machine learning to work for new tasks, reducing the dependence on large amounts of labeled data, and so on. In this talk I survey how adversarial techniques in machine learning are involved in several of these new research frontiers.
Presented at ICLR2019

[@alliedtoasters]

Excellent talk!! Ian is smart but also a good, clear speaker - he uses well-chosen words and timely anecdotes/analogies to clarify truly complicated topics.

[@kikirizki4318]

Ian is a good fellow, be like Ian

[@nkorochinaechetam2516]

I read about him and looked him up on RUclips. Brilliance and speech is in UNION

[@GiangNguyen-of4qf]

Thanks Ian for this excellent talk.

[@blacklabelmansociety]

Amazing

[@eyalpaz3095]

Great talk

[@manishbolbanda9872]

to convert quite a blur image to a high resolution clear image which one should i use GAN or VAE?

[@NozUrbina]

Where can we get the photos from the slides?

[@samraharif7510]

Can you tell me a latest research direction on Adversarial Attacks amd defense. I am going to pursue my PhD this year and I want conduct research on this topic.

[@mohsendirbaz3908]

If your adversary constantly innovate, the whole frequentist approach collapses.

[@kicksessentials7]

The anxious vermicelli broadly tempt because surfboard cephalometrically educate at a naive violin. warlike, rich soap

[@christiansmith-of7dt]

Its super depressing

[@pattiknuth4822]

What an annoying speaker. I hate when the speaker raises the pitch of his voice at the end of each sentence. Incredibly irritating. It's distracting.