{"paper":{"title":"Hardware-Oriented Inference Complexity of Kolmogorov-Arnold Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Kolmogorov-Arnold Networks now have platform-independent formulas that count real multiplications, bit operations, and additions for hardware inference.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bilal Khalid, Jaroslaw E. Prilepsky, Pedro Freire, Sergei K. Turitsyn","submitted_at":"2026-04-03T10:18:05Z","abstract_excerpt":"Kolmogorov-Arnold Networks (KANs) have recently emerged as a powerful architecture for various machine learning applications. However, their unique structure raises significant concerns regarding their computational overhead. Existing studies primarily evaluate KAN complexity in terms of Floating-Point Operations (FLOPs) required for GPU-based training and inference. However, in many latency-sensitive and power-constrained deployment scenarios, such as neural network-driven non-linearity mitigation in optical communications or channel state estimation in wireless communications, training is pe"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"we derive generalized, platform-independent formulae for evaluating the hardware inference complexity of KANs in terms of Real Multiplications (RM), Bit Operations (BOP), and Number of Additions and Bit-Shifts (NABS). We extend our analysis across multiple KAN variants, including B-spline, Gaussian Radial Basis Function (GRBF), Chebyshev, and Fourier KANs.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That RM, BOP, and NABS counts derived solely from network structure accurately predict real hardware resource use and latency without additional factors such as memory access patterns or routing overhead.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Derives generalized formulas for KAN inference complexity using RM, BOP, and NABS metrics across B-spline, GRBF, Chebyshev, and Fourier variants.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Kolmogorov-Arnold Networks now have platform-independent formulas that count real multiplications, bit operations, and additions for hardware inference.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"4b5b1543c135bdc0db32f62b924ce1806757a6bdaac506b649598bb8a8fef406"},"source":{"id":"2604.03345","kind":"arxiv","version":2},"verdict":{"id":"3d817662-71e4-4a19-8846-f6d7c3672a7f","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-13T20:26:04.678464Z","strongest_claim":"we derive generalized, platform-independent formulae for evaluating the hardware inference complexity of KANs in terms of Real Multiplications (RM), Bit Operations (BOP), and Number of Additions and Bit-Shifts (NABS). We extend our analysis across multiple KAN variants, including B-spline, Gaussian Radial Basis Function (GRBF), Chebyshev, and Fourier KANs.","one_line_summary":"Derives generalized formulas for KAN inference complexity using RM, BOP, and NABS metrics across B-spline, GRBF, Chebyshev, and Fourier variants.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That RM, BOP, and NABS counts derived solely from network structure accurately predict real hardware resource use and latency without additional factors such as memory access patterns or routing overhead.","pith_extraction_headline":"Kolmogorov-Arnold Networks now have platform-independent formulas that count real multiplications, bit operations, and additions for hardware inference."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.03345/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"}