{"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/graph-neural-networks-with-configuration","title":"Graph neural networks with configuration cross-attention for tensor compilers","arxiv_id":"2405.16623","date":"2024-05-26","proceeding":null,"authors":["Dmitrii Khizbullin","Eduardo Rocha de Andrade","Thanh Hau Nguyen","Matheus Pedroza Ferreira","David R. Pugh"],"abstract":"With the recent popularity of neural networks comes the need for efficient serving of inference workloads. A neural network inference workload can be represented as a computational graph with nodes as operators transforming multidimensional tensors. The tensors can be transposed and/or tiled in a combinatorially large number of ways, some configurations leading to accelerated inference. We propose TGraph, a neural graph architecture that allows screening for fast configurations of the target computational graph, thus representing an artificial intelligence (AI) tensor compiler in contrast to the traditional heuristics-based compilers. The proposed solution improves mean Kendall's $\\tau$ across layout collections of TpuGraphs from 29.8% of the reliable baseline to 67.4% of TGraph. We estimate the potential CO$_2$ emission reduction associated with our work to be equivalent to over 50% of the total household emissions in the areas hosting AI-oriented data centers.","url_abs":"https://arxiv.org/abs/2405.16623v2","url_pdf":"https://arxiv.org/pdf/2405.16623v2.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":"graph-neural-networks-with-configuration","repo_url":"https://github.com/thanhhau097/google_fast_or_slow","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"graph-property-prediction","task_name":"Graph Property Prediction"},{"task_slug":"graph-regression","task_name":"Graph Regression"},{"task_slug":"runtime-ranking","task_name":"Runtime ranking"}],"methods":[{"method_slug":"graphsage","method_name":"GraphSAGE"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/runtime-ranking-on-tpugraphs-layout-mean","task":"Runtime ranking","dataset":"TpuGraphs Layout mean","model":"TGraph","rank_in_archive_order":1,"of":2,"metrics":{"Kendall's Tau":"0.674"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}