DiscoverAI, Radiology and Ethics: A Podcast SeriesAvoiding Shortcut Solutions in Machine Learning Models
Avoiding Shortcut Solutions in Machine Learning Models

Avoiding Shortcut Solutions in Machine Learning Models

Update: 2022-03-14
Share

Description

In this podcast Joshua Robinson discusses his work at MIT and his recent, lead author paper on how contrastive learning might lead to more reliable predictions in AI. Josh’s paper is at the NeurIPS proceedings website:


https://papers.nips.cc/paper/2021/hash/27934a1f19d678a1377c257b9a780e80-Abstract.html.

Comments 
00:00
00:00
x

0.5x

0.8x

1.0x

1.25x

1.5x

2.0x

3.0x

Sleep Timer

Off

End of Episode

5 Minutes

10 Minutes

15 Minutes

30 Minutes

45 Minutes

60 Minutes

120 Minutes

Avoiding Shortcut Solutions in Machine Learning Models

Avoiding Shortcut Solutions in Machine Learning Models

Emory University Center for Ethics