Litcius/Paper detail

Greater than the Sum of its LUTs: Scaling Up LUT-based Neural Networks with AmigoLUT

Olivia Weng, Marta Andronic, Danial Zuberi, Jiaqing Chen, Caleb Geniesse, George A. Constantinides, Nhan Viet Tran, Nicholas J. Fraser, J. Duarte, Ryan Kastner

202514 citationsDOIOpen Access PDF

Abstract

Applications like high-energy physics and cybersecurity require extremely high throughput and low latency neural network (NN) inference. Lookup-table-based NNs address these constraints by implementing NNs as lookup tables (LUTs), achieving inference latency on the order of nanoseconds. Since LUTs are a fundamental FPGA building block, LUT-based NNs efficiently map to FPGAs. LogicNets (and its successors) form one class of LUT-based NNs that target FPGAs, mapping neurons directly to LUTs to meet low latency constraints with minimal resources. However, it is difficult to build larger, more performant LUT-based NNs like LogicNets because LUT usage increases exponentially with respect to neuron fan-in (i.e., number of synapses X synapse bitwidth). A large LUT-based NN quickly runs out of LUTs on an FPGA. Our work AmigoLUT addresses this issue by creating ensembles of smaller LUT-based NNs that scale linearly with respect to the number of models. AmigoLUT improves the scalability of LUT-based NNs, reaching higher throughput with up to an order of magnitude fewer LUTs than the largest LUT-based NNs.

Topics & Concepts

Lookup tableArtificial neural networkComputer scienceScalingArtificial intelligenceMathematicsProgramming languageGeometryMachine Learning and Data ClassificationNeural Networks and ApplicationsAnomaly Detection Techniques and Applications
Greater than the Sum of its LUTs: Scaling Up LUT-based Neural Networks with AmigoLUT | Litcius