| Date |
Speaker(s) |
Title |
| 1/3 |
Chaitanya Ryali |
Self-supervised learning with
background augmentations
|
| 1/10 |
Fatemehsadat Mireshghallah |
Style pooling: automatic text style obfuscation for improved classification fairness |
| 1/17
| Martin Luther King Jr. Day |
I Have A Dream |
| 1/24
| Lawrence Saul |
A nonlinear matrix decomposition for
mining the zeros of sparse data |
| 1/31 |
Daniella Horan
(HUJI) |
When is unsupervised disentanglement possible? |
| 2/7 |
Bodhisattwa
Majumder |
Producing explanations with
commonsense and interactions |
| 2/14 |
Rui Wang
|
Physics-guided deep learning
for dynamics forecasting
|
| 2/21 |
Presidents Day
|
Farewell Address
Gettysburg Address |
| 2/28 |
Yamini Bansal
(Harvard) |
Understanding representations
in deep learning |
| 3/7 |
Yixin Wang
(UMich)
|
Posterior collapse and latent
variable non-identifiability |