{"id":593,"date":"2024-12-16T07:04:11","date_gmt":"2024-12-16T07:04:11","guid":{"rendered":"https:\/\/mailitics.com\/index.php\/2024\/12\/16\/2412-09687\/"},"modified":"2024-12-16T07:04:11","modified_gmt":"2024-12-16T07:04:11","slug":"2412-09687","status":"publish","type":"post","link":"https:\/\/mailitics.com\/index.php\/2024\/12\/16\/2412-09687\/","title":{"rendered":"DQA: An Efficient Method for Deep Quantization of Deep Neural Network Activations"},"content":{"rendered":"

DQA: An Efficient Method for Deep Quantization of Deep Neural Network Activations
\n \t

\n
<\/BR>
\n
\n \t

\n
<\/BR><\/p>\n

arXiv:2412.09687v1 Announce Type: cross
\nAbstract: Quantization of Deep Neural Network (DNN) activations is a commonly used technique to reduce compute and memory demands during DNN inference, which can be particularly beneficial on resource-constrained devices. To achieve high accuracy, existing methods for quantizing activations rely on complex mathematical computations or perform extensive searches for the best hyper-parameters. However, these expensive operations are impractical on devices with limited computation capabilities, memory capacities, and energy budgets. Furthermore, many existing methods do not focus on sub-6-bit (or deep) quantization.
\n To fill these gaps, in this paper we propose DQA (Deep Quantization of DNN Activations), a new method that focuses on sub-6-bit quantization of activations and leverages simple shifting-based operations and Huffman coding to be efficient and achieve high accuracy. We evaluate DQA with 3, 4, and 5-bit quantization levels and three different DNN models for two different tasks, image classification and image segmentation, on two different datasets. DQA shows significantly better accuracy (up to 29.28%) compared to the direct quantization method and the state-of-the-art NoisyQuant for sub-6-bit quantization.<\/div>\n

\t

\n
<\/BR>
\n Wenhao Hu, Paul Henderson, Jos’e Cano
\n \t

\n
<\/BR>
\nGo to original source<\/a>
\n \t

\n
<\/BR><\/p>\n","protected":false},"excerpt":{"rendered":"

DQA: An Efficient Method for Deep Quantization of Deep Neural Network Activations arXiv:2412.09687v1 Announce Type: cross Abstract: Quantization of Deep Neural Network (DNN) activations is a commonly used technique to reduce compute and memory demands during DNN inference, which can be particularly beneficial on resource-constrained devices. To achieve high accuracy, existing methods for quantizing activations […]<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[62,611,113,112],"tags":[714,4,713],"class_list":["post-593","post","type-post","status-publish","format-standard","hentry","category-aimldsaimlds","category-cs-cv","category-cs-lg","category-stat-ml","tag-activations","tag-deep","tag-quantization"],"_links":{"self":[{"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/posts\/593"}],"collection":[{"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/comments?post=593"}],"version-history":[{"count":0,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/posts\/593\/revisions"}],"wp:attachment":[{"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/media?parent=593"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/categories?post=593"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/tags?post=593"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}