Panel on Machine Learning
Apr. 24th, 2007 01:59 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
guslPanel on Machine Learning at lunchtime:
They seemed to be going on about an old question on the philosophy of science: should we care about engineering or science? (or both?), i.e, the trade-off between explanatory unification and empirical success. The interesting cases are the ones where there is a mix, in which we have partial unification.
Carlos talked about Query-specific learning.
Bob Murphy talked about global models of the cell.
Feinberg: trying to answer open-world forecasting questions. Statistics is not enough, but how can we integrate that information with say, newspaper headlines? He claimed that Machine Learning community didn't do work on sampling, experimental design, causal inference, etc.
I liked the format. While the superficial nature of the presentations (12 minutes each) led to some silly shallow disagreements, it inspires interesting debates among the audience.
They seemed to be going on about an old question on the philosophy of science: should we care about engineering or science? (or both?), i.e, the trade-off between explanatory unification and empirical success. The interesting cases are the ones where there is a mix, in which we have partial unification.
Carlos talked about Query-specific learning.
Bob Murphy talked about global models of the cell.
Feinberg: trying to answer open-world forecasting questions. Statistics is not enough, but how can we integrate that information with say, newspaper headlines? He claimed that Machine Learning community didn't do work on sampling, experimental design, causal inference, etc.
I liked the format. While the superficial nature of the presentations (12 minutes each) led to some silly shallow disagreements, it inspires interesting debates among the audience.