{"url":"/method/logan","slug":"logan","name":"LOGAN","full_name":"LOGAN","full_name_withheld":false,"description_markdown":"**LOGAN** is a generative adversarial network that uses a latent optimization approach using [natural gradient descent](https://paperswithcode.com/method/natural-gradient-descent) (NGD). For the Fisher matrix in NGD, the authors use the empirical Fisher $F'$ with Tikhonov damping:\r\n\r\n$$ F' = g \\cdot g^{T} + \\beta{I} $$\r\n\r\nThey also use Euclidian Norm regularization for the optimization step.\r\n\r\nFor LOGAN's base architecture, [BigGAN-deep](https://paperswithcode.com/method/biggan-deep) is used with a few modifications: increasing the size of the latent source from $186$ to $256$, to compensate the randomness of the source lost\r\nwhen optimising $z$. 2, using the uniform distribution $U\\left(−1, 1\\right)$ instead of the standard normal distribution $N\\left(0, 1\\right)$ for $p\\left(z\\right)$ to be consistent with the clipping operation, using  leaky [ReLU](https://paperswithcode.com/method/relu) (with the slope of 0.2 for the negative part) instead of ReLU as the non-linearity for smoother gradient flow for $\\frac{\\delta{f}\\left(z\\right)}{\\delta{z}}$ .","description_state":"present","introduced_year":null,"introduced_by":{"title":"LOGAN: Latent Optimisation for Generative Adversarial Networks","paper":"/paper/logan-latent-optimisation-for-generative-1","first_author":"Yan Wu","n_authors":5,"url_abs":null,"archive_paper_url":"https://paperswithcode.com/paper/logan-latent-optimisation-for-generative-1"},"source":{"url":"https://arxiv.org/abs/1912.00953v2","title":"LOGAN: Latent Optimisation for Generative Adversarial Networks","url_on_a_paper_host":true},"code_snippet_url":"https://github.com/Hosein47/LOGAN","code_snippet_url_on_a_code_host":true,"categories":[{"area":"Computer Vision","area_id":"computer-vision","collection":"Generative Adversarial Networks","url":"/methods/category/generative-adversarial-networks","pwc_aliases":[]},{"area":"Computer Vision","area_id":"computer-vision","collection":"Generative Models","url":"/methods/category/generative-models","pwc_aliases":[]}],"n_papers_tagged":6,"archive_num_papers":6,"papers_newest_first":[{"paper":null,"title":"Auditing Algorithmic Fairness in Machine Learning for Health with Severity-Based LOGAN","date":"2022-11-16","arxiv_id":"2211.08742","n_code_links":0,"syntology":null},{"paper":null,"title":"Sinogram Denoise Based on Generative Adversarial Networks","date":"2021-08-09","arxiv_id":"2108.03903","n_code_links":0,"syntology":null},{"paper":null,"title":"Direct Reconstruction of Linear Parametric Images from Dynamic PET Using Nonlocal Deep Image Prior","date":"2021-06-18","arxiv_id":"2106.10359","n_code_links":0,"syntology":null},{"paper":"/paper/logan-local-group-bias-detection-by","title":"LOGAN: Local Group Bias Detection by Clustering","date":"2020-10-06","arxiv_id":"2010.02867","n_code_links":1,"syntology":null},{"paper":null,"title":"Allpass Feedback Delay Networks","date":"2020-07-14","arxiv_id":"2007.07337","n_code_links":0,"syntology":null},{"paper":"/paper/logan-latent-optimisation-for-generative-1","title":"LOGAN: Latent Optimisation for Generative Adversarial Networks","date":"2019-12-02","arxiv_id":"1912.00953","n_code_links":1,"syntology":{"ran":2,"of":8,"unverified":6,"pointer_only":0}}],"papers_shown":6,"tasks":[{"task":"/task/bias-detection","name":"Bias Detection","papers":2},{"task":"/task/clustering","name":"Clustering","papers":2},{"task":"/task/machine-learning","name":"BIG-bench Machine Learning","papers":1},{"task":"/task/computational-efficiency","name":"Computational Efficiency","papers":1},{"task":"/task/conditional-image-generation","name":"Conditional Image Generation","papers":1},{"task":"/task/denoising","name":"Denoising","papers":1},{"task":"/task/fairness","name":"Fairness","papers":1},{"task":"/task/image-generation","name":"Image Generation","papers":1}],"tasks_shown":8,"n_tasks":8,"usage_by_year":[{"year":"2019","papers":1},{"year":"2020","papers":2},{"year":"2021","papers":2},{"year":"2022","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/logan"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}