Abstract | ||
---|---|---|
We train multi-task (variational) autoencoders on linguistic tasks and analyze the learned hidden sentence representations. The representations change significantly when translation and part-of-speech decoders are added. The more decoders are attached, the better the models cluster sentences according to their syntactic similarity, as the representation space becomes less entangled. We compare standard unconstrained autoencoders to variational autoencoders and find significant differences. We achieve better disentanglement with the standard autoencoder, which goes against recent work on variational autoencoders in the visual domain. |
Year | DOI | Venue |
---|---|---|
2018 | 10.1007/978-3-319-97982-3_13 | ADVANCES IN COMPUTATIONAL INTELLIGENCE SYSTEMS (UKCI) |
Keywords | Field | DocType |
NLP,Variational,Autoencoder,Disentanglement,Representation learning,Syntax | Autoencoder,Computer science,Natural language processing,Artificial intelligence,Syntax,Sentence,Feature learning | Conference |
Volume | ISSN | Citations |
840 | 2194-5357 | 0 |
PageRank | References | Authors |
0.34 | 5 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Gino Brunner | 1 | 20 | 5.76 |
Yuyi Wang | 2 | 22 | 10.01 |
Rogert Wattenhofer | 3 | 5580 | 384.89 |
Michael Weigelt | 4 | 2 | 0.69 |