{"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/edge-moe-memory-efficient-multi-task-vision","title":"Edge-MoE: Memory-Efficient Multi-Task Vision Transformer Architecture with Task-level Sparsity via Mixture-of-Experts","arxiv_id":"2305.18691","date":"2023-05-30","proceeding":null,"authors":["Rishov Sarkar","Hanxue Liang","Zhiwen Fan","Zhangyang Wang","Cong Hao"],"abstract":"Computer vision researchers are embracing two promising paradigms: Vision Transformers (ViTs) and Multi-task Learning (MTL), which both show great performance but are computation-intensive, given the quadratic complexity of self-attention in ViT and the need to activate an entire large MTL model for one task. M$^3$ViT is the latest multi-task ViT model that introduces mixture-of-experts (MoE), where only a small portion of subnetworks (\"experts\") are sparsely and dynamically activated based on the current task. M$^3$ViT achieves better accuracy and over 80% computation reduction but leaves challenges for efficient deployment on FPGA. Our work, dubbed Edge-MoE, solves the challenges to introduce the first end-to-end FPGA accelerator for multi-task ViT with a collection of architectural innovations, including (1) a novel reordering mechanism for self-attention, which requires only constant bandwidth regardless of the target parallelism; (2) a fast single-pass softmax approximation; (3) an accurate and low-cost GELU approximation; (4) a unified and flexible computing unit that is shared by almost all computational layers to maximally reduce resource usage; and (5) uniquely for M$^3$ViT, a novel patch reordering method to eliminate memory access overhead. Edge-MoE achieves 2.24x and 4.90x better energy efficiency comparing with GPU and CPU, respectively. A real-time video demonstration is available online, along with our open-source code written using High-Level Synthesis.","url_abs":"https://arxiv.org/abs/2305.18691v2","url_pdf":"https://arxiv.org/pdf/2305.18691v2.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":"edge-moe-memory-efficient-multi-task-vision","repo_url":"https://github.com/sharc-lab/edge-moe","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":null,"task_name":"CPU"},{"task_slug":null,"task_name":"GPU"},{"task_slug":"high-level-synthesis","task_name":"High-Level Synthesis"},{"task_slug":"mixture-of-experts","task_name":"Mixture-of-Experts"},{"task_slug":"multi-task-learning","task_name":"Multi-Task Learning"}],"methods":[{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2305.18691","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}