[0:05] Plan Day
[0:48] Think about product distribution case for semi-adversarial setting
[0:44] Write up product distribution case
[0:19] Improve de-serialization speed for risk estimation code
[0:48] Set up 2 more datasets for risk estimation project
[0:34] Understand where estimates are off and why
Possible reasons I discovered:
-very narrow set of words get classified as anchors (causes distributional skew, often leads to under-estimate of risk)
-very easy and common examples end up not getting classified as anchors (leads to over-estimate of risk)
-removing the anchor from the prediction can be a large hit to accuracy if classification is based on only a few features (leads to over-estimate of risk)
[0:10] Think about how to improve the estimates
[0:15] Look over Uri / Frederik’s papers on representation learning for counterfactual estimation
[0:42] More reading about non-negative matrix factorization
[1:05] Meeting about code translation project
[1:00] Climbing


Back after a long hiatus. Decided to start doing these again because my productivity / fitness were slipping over the summer.

[0:17] Plan Day
[1:32] Think about semi-adversarial setting
[0:38] Write about semi-adversarial setting
[1:34] Set up risk estimation experiments
[0:55] Lifting / cardio
squat 3×5@125
bench press 3×5@105
stationary bike 4x(60s jog/30s sprint)
[0:45] Read about non-negative matrix factorization
[1:15] Skype call
[2:15] E-mails 😦
[0:10] Think about PCFG approach
[0:10] E-mail collaborators about PCFG approach