{"url":"/method/filter-response-normalization","slug":"filter-response-normalization","name":"Filter Response Normalization","full_name":"Filter Response Normalization","full_name_withheld":false,"description_markdown":"**Filter Response Normalization (FRN)** is a type of normalization that combines normalization and an activation function, which can be used as a replacement for other normalizations and activations. It operates on each activation channel of each batch element independently, eliminating the dependency on other batch elements. \r\n\r\nTo demonstrate, assume we are dealing with the feed-forward convolutional neural network. We follow the usual convention that the filter responses (activation maps) produced after a [convolution](https://paperswithcode.com/method/convolution) operation are a [4D ](https://paperswithcode.com/method/4d-a)tensor $X$ with shape $[B, W, H, C]$, where $B$ is the mini-batch size, $W, H$ are the spatial extents of the map, and $C$ is the number of filters used in convolution. $C$ is also referred to as output channels. Let $x = X_{b,:,:,c} \\in \\mathcal{R}^{N}$, where $N = W \\times H$, be the vector of filter responses for the $c^{th}$ filter for the $b^{th}$ batch point. \r\nLet $\\nu^2 = \\sum\\_i x_i^2/N$, be the mean squared norm of $x$. \r\n\r\nThen Filter Response Normalization is defined as the following:\r\n\r\n$$\r\n\\hat{x} = \\frac{x}{\\sqrt{\\nu^2 + \\epsilon}},\r\n$$\r\n\r\nwhere $\\epsilon$ is a small positive constant to prevent division by zero.  \r\n\r\nA lack of mean centering in FRN can lead to activations having an arbitrary bias away from zero. Such a bias in conjunction with [ReLU](https://paperswithcode.com/method/relu) can have a detrimental effect on learning and lead to poor performance and dead units. To address this the authors augment ReLU with a learned threshold $\\tau$ to yield:\r\n\r\n$$\r\nz = \\max(y, \\tau)\r\n$$\r\n\r\nSince $\\max(y, \\tau){=}\\max(y-\\tau,0){+}\\tau{=}\\text{ReLU}{(y{-}\\tau)}{+}\\tau$, the effect of this activation is the same as having a shared bias before and after ReLU.","description_state":"present","introduced_year":null,"introduced_by":{"title":"Filter Response Normalization Layer: Eliminating Batch Dependence in the Training of Deep Neural Networks","paper":"/paper/filter-response-normalization-layer","first_author":"Saurabh Singh","n_authors":2,"url_abs":null,"archive_paper_url":"https://paperswithcode.com/paper/filter-response-normalization-layer"},"source":{"url":"https://arxiv.org/abs/1911.09737v2","title":"Filter Response Normalization Layer: Eliminating Batch Dependence in the Training of Deep Neural Networks","url_on_a_paper_host":true},"code_snippet_url":"https://github.com/gupta-abhay/pytorch-frn/blob/cc86a984fcbae61431ed41f37695e78f5e4b196e/frn.py#L9","code_snippet_url_on_a_code_host":true,"categories":[{"area":"General","area_id":"general","collection":"Normalization","url":"/methods/category/normalization","pwc_aliases":[]}],"n_papers_tagged":2,"archive_num_papers":2,"papers_newest_first":[{"paper":"/paper/deceiving-computers-in-reverse-turing-test","title":"Deceiving computers in Reverse Turing Test through Deep Learning","date":"2020-06-01","arxiv_id":"2006.11373","n_code_links":2,"syntology":null},{"paper":"/paper/filter-response-normalization-layer","title":"Filter Response Normalization Layer: Eliminating Batch Dependence in the Training of Deep Neural Networks","date":"2019-11-21","arxiv_id":"1911.09737","n_code_links":16,"syntology":{"ran":2,"of":2,"unverified":0,"pointer_only":0}}],"papers_shown":2,"tasks":[{"task":null,"name":"CAPTCHA Detection","papers":1},{"task":"/task/deep-learning","name":"Deep Learning","papers":1},{"task":"/task/image-classification","name":"Image Classification","papers":1},{"task":"/task/object-detection","name":"Object Detection","papers":1},{"task":"/task/object-detection-1","name":"object-detection","papers":1}],"tasks_shown":5,"n_tasks":5,"usage_by_year":[{"year":"2019","papers":1},{"year":"2020","papers":1}],"row_source":"methods_table","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/filter-response-normalization"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}