{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/recurrent-neural-networks-hardware","title":"Recurrent Neural Networks Hardware Implementation on FPGA","arxiv_id":"1511.05552","date":"2015-11-17","proceeding":null,"authors":["Andre Xian Ming Chang","Berin Martini","Eugenio Culurciello"],"abstract":"Recurrent Neural Networks (RNNs) have the ability to retain memory and learn\ndata sequences. Due to the recurrent nature of RNNs, it is sometimes hard to\nparallelize all its computations on conventional hardware. CPUs do not\ncurrently offer large parallelism, while GPUs offer limited parallelism due to\nsequential components of RNN models. In this paper we present a hardware\nimplementation of Long-Short Term Memory (LSTM) recurrent network on the\nprogrammable logic Zynq 7020 FPGA from Xilinx. We implemented a RNN with $2$\nlayers and $128$ hidden units in hardware and it has been tested using a\ncharacter level language model. The implementation is more than $21\\times$\nfaster than the ARM CPU embedded on the Zynq 7020 FPGA. This work can\npotentially evolve to a RNN co-processor for future mobile devices.","url_abs":"http://arxiv.org/abs/1511.05552v4","url_pdf":"http://arxiv.org/pdf/1511.05552v4.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"recurrent-neural-networks-hardware","repo_url":"https://github.com/karpathy/char-rnn","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"torch","reach":{"status":"ok"}}],"tasks":[{"task_slug":null,"task_name":"CPU"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1511.05552","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}