This presentation will review the strengths and weaknesses of using pre-trained word embeddings, and demonstrate how to incorporate more complex semantic representation schemes such as Semantic Role Labeling, AMR and SDP into your applications.
30. Iyyer and collaborators broke the tree-structured
bidirectional LSTM sentiment classification model.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40. The tale of Mr. Morton gives a great intro to
Subject predicate structure. What ever the
predicate says he does.
Source https://people.eecs.berkeley.edu/~klein/cs294-7/SP07%20cs294%20lecture%2019%20--
%20compositional%20semantics%20(6pp).pdf and https://web.stanford.edu/~jurafsky/slp3/22.pdf
41. The tale of Mr. Morton gives a great intro to
Subject predicate structure. What ever the
predicate says he does.
Source https://people.eecs.berkeley.edu/~klein/cs294-7/SP07%20cs294%20lecture%2019%20--
%20compositional%20semantics%20(6pp).pdf and https://web.stanford.edu/~jurafsky/slp3/22.pdf
42. The tale of Mr. Morton gives a great intro to
Subject predicate structure. What ever the
predicate says he does.
Source https://people.eecs.berkeley.edu/~klein/cs294-7/SP07%20cs294%20lecture%2019%20--
%20compositional%20semantics%20(6pp).pdf and https://web.stanford.edu/~jurafsky/slp3/22.pdf
43. The tale of Mr. Morton gives a great intro to
Subject predicate structure. What ever the
predicate says he does.
Source https://people.eecs.berkeley.edu/~klein/cs294-7/SP07%20cs294%20lecture%2019%20--
%20compositional%20semantics%20(6pp).pdf and https://web.stanford.edu/~jurafsky/slp3/22.pdf
54. Prepare
Data
Register and
Manage Model
Train & Test
Model
Build
Image
…
Build model
(your favorite
IDE)
Deploy
Service
Monitor
Model
Prepare Experiment Deploy
Machine Learning on Azure