Log 06-17-2013

Forget exactly what happened today. Basically, I realized that the top-down version of beam search (also referred to here earlier as loss-sensitive EP) that I had in mind was just regular EP, and that Dan Klein had done something very similar already. I then switched to thinking about another idea based on using intermediate beams to train during the learning phase. (This turns out to also not be that great of an idea, but I don’t realize this until sometime tomorrow, at which I switch to yet another idea of using an EP-like parsing method to gain better coverage, in a way very similar to my NIPS paper.)

EDIT: Oh wait, I found my log, here it is:

1. (0:15) Call about fixing car
2. (0:20) Set up dentist appointment
3. (0:15) Order new phone battery
4. (2:20)  Read Dan Klein papers
5. (1:15) Find good reading group material

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s