{"url":"/method/qhadam","slug":"qhadam","name":"QHAdam","full_name":"QHAdam","full_name_withheld":false,"description_markdown":"The **Quasi-Hyperbolic Momentum Algorithm (QHM)** is a simple alteration of [momentum SGD](https://paperswithcode.com/method/sgd-with-momentum), averaging a plain [SGD](https://paperswithcode.com/method/sgd) step with a momentum step. **QHAdam** is a QH augmented version of [Adam](https://paperswithcode.com/method/adam), where we replace both of Adam's moment estimators with quasi-hyperbolic terms. QHAdam decouples the momentum term from the current gradient when updating the weights, and decouples the mean squared gradients term from the current squared gradient when updating the weights. \r\n\r\nIn essence, it is a weighted average of the momentum and plain SGD, weighting the current gradient with an immediate discount factor $v\\_{1}$ divided by a weighted average of the mean squared gradients and the current squared gradient, weighting the current squared gradient with an immediate discount factor $v\\_{2}$. \r\n\r\n$$ \\theta\\_{t+1, i} = \\theta\\_{t, i} - \\eta\\left[\\frac{\\left(1-v\\_{1}\\right)\\cdot{g\\_{t}} + v\\_{1}\\cdot\\hat{m}\\_{t}}{\\sqrt{\\left(1-v\\_{2}\\right)g^{2}\\_{t} + v\\_{2}\\cdot{\\hat{v}\\_{t}}} + \\epsilon}\\right], \\forall{t} $$\r\n\r\nIt is recommended to set $v\\_{2} = 1$ and $\\beta\\_{2}$ same as in Adam.","description_state":"present","introduced_year":null,"introduced_by":{"title":"Quasi-hyperbolic momentum and Adam for deep learning","paper":"/paper/quasi-hyperbolic-momentum-and-adam-for-deep","first_author":"Jerry Ma","n_authors":2,"url_abs":null,"archive_paper_url":"https://paperswithcode.com/paper/quasi-hyperbolic-momentum-and-adam-for-deep"},"source":{"url":"http://arxiv.org/abs/1810.06801v4","title":"Quasi-hyperbolic momentum and Adam for deep learning","url_on_a_paper_host":true},"code_snippet_url":"https://github.com/jettify/pytorch-optimizer/blob/155246597d66dd774156599be0f07a8c6f7758aa/torch_optimizer/qhadam.py#L10","code_snippet_url_on_a_code_host":true,"categories":[{"area":"General","area_id":"general","collection":"Stochastic Optimization","url":"/methods/category/stochastic-optimization","pwc_aliases":[]}],"n_papers_tagged":1,"archive_num_papers":1,"papers_newest_first":[{"paper":"/paper/quasi-hyperbolic-momentum-and-adam-for-deep","title":"Quasi-hyperbolic momentum and Adam for deep learning","date":"2018-10-16","arxiv_id":"1810.06801","n_code_links":2,"syntology":{"ran":0,"of":7,"unverified":7,"pointer_only":7}}],"papers_shown":1,"tasks":[{"task":"/task/deep-learning","name":"Deep Learning","papers":1},{"task":"/task/stochastic-optimization","name":"Stochastic Optimization","papers":1}],"tasks_shown":2,"n_tasks":2,"usage_by_year":[{"year":"2018","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/qhadam"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}