When the state of the art is ahead of the state of understanding: unintuitive properties of deep neural networks
Fuente:
Redalyc
Saved in:
| Main Author: | |
|---|---|
| Format: | Artículo científico |
| Language: | en |
| Published: |
Universitat de València
2019
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1876456401524490240 |
|---|---|
| author | Joan Serrà |
| author_facet | Joan Serrà |
| contents | When the state of the art is ahead of the state of understanding: unintuitive properties of deep neural networks Joan Serrà Comunicación Deep learning neural networks machine learning unintuitive properties Deep learning is an undeniably hot topic, not only within both academia and industry, but also among society and the media. The reasons for the advent of its popularity are manifold: unprecedented availability of data and computing power, some innovative methodologies, minor but significant technical tricks, etc. However, interestingly, the current success and practice of deep learning seems to be uncorrelated with its theoretical, more formal understanding. And with that, deep learning’s state-of-the-art presents a number of unintuitive properties or situations. In this note, I highlight some of these unintuitive properties, trying to show relevant recent work, and expose the need to get insight into them, either by formal or more empirical means. 2019 artículo científico 2174-3487 https://www.redalyc.org/articulo.oa?id=511766773021 https://www.redalyc.org/journal/5117/511766773021/ https://www.redalyc.org/journal/5117/511766773021/html/ https://www.redalyc.org/journal/5117/511766773021/511766773021.epub https://www.redalyc.org/journal/5117/511766773021/movil https://doi.org/10.7203/metode.9.11035 en http://www.redalyc.org/revista.oa?id=5117 Mètode Science Studies Journal application/pdf Universitat de València Mètode Science Studies Journal (España) Vol.9 |
| format | Artículo científico |
| id | redalyc_511766773021 |
| institution | Redalyc |
| language | en |
| publishDate | 2019 |
| publisher | Universitat de València |
| spellingShingle | When the state of the art is ahead of the state of understanding: unintuitive properties of deep neural networks Joan Serrà Comunicación Deep learning neural networks machine learning unintuitive properties When the state of the art is ahead of the state of understanding: unintuitive properties of deep neural networks Joan Serrà Comunicación Deep learning neural networks machine learning unintuitive properties Deep learning is an undeniably hot topic, not only within both academia and industry, but also among society and the media. The reasons for the advent of its popularity are manifold: unprecedented availability of data and computing power, some innovative methodologies, minor but significant technical tricks, etc. However, interestingly, the current success and practice of deep learning seems to be uncorrelated with its theoretical, more formal understanding. And with that, deep learning’s state-of-the-art presents a number of unintuitive properties or situations. In this note, I highlight some of these unintuitive properties, trying to show relevant recent work, and expose the need to get insight into them, either by formal or more empirical means. 2019 artículo científico 2174-3487 https://www.redalyc.org/articulo.oa?id=511766773021 https://www.redalyc.org/journal/5117/511766773021/ https://www.redalyc.org/journal/5117/511766773021/html/ https://www.redalyc.org/journal/5117/511766773021/511766773021.epub https://www.redalyc.org/journal/5117/511766773021/movil https://doi.org/10.7203/metode.9.11035 en http://www.redalyc.org/revista.oa?id=5117 Mètode Science Studies Journal application/pdf Universitat de València Mètode Science Studies Journal (España) Vol.9 |
| title | When the state of the art is ahead of the state of understanding: unintuitive properties of deep neural networks |
| topic | Comunicación Deep learning neural networks machine learning unintuitive properties |
| url | https://www.redalyc.org/articulo.oa?id=511766773021 https://www.redalyc.org/journal/5117/511766773021/ https://www.redalyc.org/journal/5117/511766773021/html/ https://www.redalyc.org/journal/5117/511766773021/511766773021.epub https://www.redalyc.org/journal/5117/511766773021/movil https://doi.org/10.7203/metode.9.11035 |