Title: Question about ReLU in Multi-Head Attention · Issue #21 · DeepGraphLearning/RecommenderSystems · GitHub
Open Graph Title: Question about ReLU in Multi-Head Attention · Issue #21 · DeepGraphLearning/RecommenderSystems
X Title: Question about ReLU in Multi-Head Attention · Issue #21 · DeepGraphLearning/RecommenderSystems
Description: In multi-head attention, there is a relu after queries, keys, and values. Is this a correct implementation? The paper did not mention the relu in Eq. 5. Besides, it seems that the relu will make the attention matrix always positive. # Li...
Open Graph Description: In multi-head attention, there is a relu after queries, keys, and values. Is this a correct implementation? The paper did not mention the relu in Eq. 5. Besides, it seems that the relu will make th...
X Description: In multi-head attention, there is a relu after queries, keys, and values. Is this a correct implementation? The paper did not mention the relu in Eq. 5. Besides, it seems that the relu will make th...
Opengraph URL: https://github.com/DeepGraphLearning/RecommenderSystems/issues/21
X: @github
Domain: patch-diff.githubusercontent.com
{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"Question about ReLU in Multi-Head Attention","articleBody":"In multi-head attention, there is a relu after queries, keys, and values. Is this a correct implementation? The paper did not mention the relu in Eq. 5. Besides, it seems that the relu will make the attention matrix always positive.\r\n\r\n```python\r\n# Linear projections\r\nQ = tf.layers.dense(queries, num_units, activation=tf.nn.relu)\r\nK = tf.layers.dense(keys, num_units, activation=tf.nn.relu)\r\nV = tf.layers.dense(values, num_units, activation=tf.nn.relu)```","author":{"url":"https://github.com/ArtanisCV","@type":"Person","name":"ArtanisCV"},"datePublished":"2022-08-10T01:38:31.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":0},"url":"https://github.com/21/RecommenderSystems/issues/21"}
| route-pattern | /_view_fragments/issues/show/:user_id/:repository/:id/issue_layout(.:format) |
| route-controller | voltron_issues_fragments |
| route-action | issue_layout |
| fetch-nonce | v2:17670a3f-f0b7-011c-529e-03ea45a75880 |
| current-catalog-service-hash | 81bb79d38c15960b92d99bca9288a9108c7a47b18f2423d0f6438c5b7bcd2114 |
| request-id | B170:1D7F:2DFAF05:3AC748F:697DDC82 |
| html-safe-nonce | b97b764c7cac87e27195b8f1ad8b1986d6994439e894c12272879c816ba043ad |
| visitor-payload | eyJyZWZlcnJlciI6IiIsInJlcXVlc3RfaWQiOiJCMTcwOjFEN0Y6MkRGQUYwNTozQUM3NDhGOjY5N0REQzgyIiwidmlzaXRvcl9pZCI6IjgwOTQzNDg4NjE2MjkxMjc4MTAiLCJyZWdpb25fZWRnZSI6ImlhZCIsInJlZ2lvbl9yZW5kZXIiOiJpYWQifQ== |
| visitor-hmac | 7185c1b1c65c4548b6032413f902138f299ef2ba296564c9bb32776c53ef3784 |
| hovercard-subject-tag | issue:1333953042 |
| github-keyboard-shortcuts | repository,issues,copilot |
| google-site-verification | Apib7-x98H0j5cPqHWwSMm6dNU4GmODRoqxLiDzdx9I |
| octolytics-url | https://collector.github.com/github/collect |
| analytics-location | / |
| fb:app_id | 1401488693436528 |
| apple-itunes-app | app-id=1477376905, app-argument=https://github.com/_view_fragments/issues/show/DeepGraphLearning/RecommenderSystems/21/issue_layout |
| twitter:image | https://opengraph.githubassets.com/aaf52aea6b47652e2b572f1f5f1b38af2755df92bae15919207c76cff0185fd2/DeepGraphLearning/RecommenderSystems/issues/21 |
| twitter:card | summary_large_image |
| og:image | https://opengraph.githubassets.com/aaf52aea6b47652e2b572f1f5f1b38af2755df92bae15919207c76cff0185fd2/DeepGraphLearning/RecommenderSystems/issues/21 |
| og:image:alt | In multi-head attention, there is a relu after queries, keys, and values. Is this a correct implementation? The paper did not mention the relu in Eq. 5. Besides, it seems that the relu will make th... |
| og:image:width | 1200 |
| og:image:height | 600 |
| og:site_name | GitHub |
| og:type | object |
| og:author:username | ArtanisCV |
| hostname | github.com |
| expected-hostname | github.com |
| None | 60279d4097367e16897439d16d6bbe4180663db828c666eeed2656988ffe59f6 |
| turbo-cache-control | no-preview |
| go-import | github.com/DeepGraphLearning/RecommenderSystems git https://github.com/DeepGraphLearning/RecommenderSystems.git |
| octolytics-dimension-user_id | 38018154 |
| octolytics-dimension-user_login | DeepGraphLearning |
| octolytics-dimension-repository_id | 167231432 |
| octolytics-dimension-repository_nwo | DeepGraphLearning/RecommenderSystems |
| octolytics-dimension-repository_public | true |
| octolytics-dimension-repository_is_fork | false |
| octolytics-dimension-repository_network_root_id | 167231432 |
| octolytics-dimension-repository_network_root_nwo | DeepGraphLearning/RecommenderSystems |
| turbo-body-classes | logged-out env-production page-responsive |
| disable-turbo | false |
| browser-stats-url | https://api.github.com/_private/browser/stats |
| browser-errors-url | https://api.github.com/_private/browser/errors |
| release | 7c85641c598ad130c74f7bcc27f58575cac69551 |
| ui-target | full |
| theme-color | #1e2327 |
| color-scheme | light dark |
Links:
Viewport: width=device-width