{"paper":{"title":"Polaris: Multi-Fidelity Design Space Exploration of Deep Learning Accelerators","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Calvin Lin, Charles Hong, Chirag Sakhuja","submitted_at":"2024-12-20T04:14:43Z","abstract_excerpt":"This paper presents a tool for automatically exploring the design space of deep learning accelerators (DLAs).\n  Our main advancement is Starlight, a data-driven performance model that uses transfer learning to bridge the gap between fast, low-fidelity evaluation methods (such as analytical models) and slow, high-fidelity evaluation methods (such as RTL simulation). Starlight is fast: It can provide 6,500 predictions per second, allowing the evaluation of millions of configurations per hour. Starlight is accurate: It predicts the energy-delay product measured by RTL simulation with 99\\% accurac"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.15548","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2412.15548/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}