Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313444588> ?p ?o ?g. }
Showing items 1 to 75 of
75
with 100 items per page.
- W4313444588 abstract "With an ever-growing number of parameters defining increasingly complex networks, Deep Learning has led to several breakthroughs surpassing human performance. As a result, data movement for these millions of model parameters causes a growing imbalance known as the memory wall. Neuromorphic computing is an emerging paradigm that confronts this imbalance by performing computations directly in analog memories. On the software side, the sequential Backpropagation algorithm prevents efficient parallelization and thus fast convergence. A novel method, Direct Feedback Alignment, resolves inherent layer dependencies by directly passing the error from the output to each layer. At the intersection of hardware/software co-design, there is a demand for developing algorithms that are tolerable to hardware nonidealities. Therefore, this work explores the interrelationship of implementing bio-plausible learning in-situ on neuromorphic hardware, emphasizing energy, area, and latency constraints. Using the benchmarking framework DNN+NeuroSim, we investigate the impact of hardware nonidealities and quantization on algorithm performance, as well as how network topologies and algorithm-level design choices can scale latency, energy and area consumption of a chip. To the best of our knowledge, this work is the first to compare the impact of different learning algorithms on Compute-In-Memory-based hardware and vice versa. The best results achieved for accuracy remain Backpropagation-based, notably when facing hardware imperfections. Direct Feedback Alignment, on the other hand, allows for significant speedup due to parallelization, reducing training time by a factor approaching N for N-layered networks." @default.
- W4313444588 created "2023-01-06" @default.
- W4313444588 creator A5007962593 @default.
- W4313444588 creator A5017567485 @default.
- W4313444588 creator A5058073627 @default.
- W4313444588 creator A5076776305 @default.
- W4313444588 creator A5080236384 @default.
- W4313444588 creator A5081246011 @default.
- W4313444588 date "2022-12-29" @default.
- W4313444588 modified "2023-09-28" @default.
- W4313444588 title "Biologically Plausible Learning on Neuromorphic Hardware Architectures" @default.
- W4313444588 doi "https://doi.org/10.48550/arxiv.2212.14337" @default.
- W4313444588 hasPublicationYear "2022" @default.
- W4313444588 type Work @default.
- W4313444588 citedByCount "0" @default.
- W4313444588 crossrefType "posted-content" @default.
- W4313444588 hasAuthorship W4313444588A5007962593 @default.
- W4313444588 hasAuthorship W4313444588A5017567485 @default.
- W4313444588 hasAuthorship W4313444588A5058073627 @default.
- W4313444588 hasAuthorship W4313444588A5076776305 @default.
- W4313444588 hasAuthorship W4313444588A5080236384 @default.
- W4313444588 hasAuthorship W4313444588A5081246011 @default.
- W4313444588 hasBestOaLocation W43134445881 @default.
- W4313444588 hasConcept C108583219 @default.
- W4313444588 hasConcept C113775141 @default.
- W4313444588 hasConcept C11413529 @default.
- W4313444588 hasConcept C118524514 @default.
- W4313444588 hasConcept C149635348 @default.
- W4313444588 hasConcept C151927369 @default.
- W4313444588 hasConcept C154945302 @default.
- W4313444588 hasConcept C155032097 @default.
- W4313444588 hasConcept C173608175 @default.
- W4313444588 hasConcept C199360897 @default.
- W4313444588 hasConcept C2777904410 @default.
- W4313444588 hasConcept C2780513914 @default.
- W4313444588 hasConcept C28855332 @default.
- W4313444588 hasConcept C41008148 @default.
- W4313444588 hasConcept C45374587 @default.
- W4313444588 hasConcept C50644808 @default.
- W4313444588 hasConcept C68339613 @default.
- W4313444588 hasConcept C9390403 @default.
- W4313444588 hasConceptScore W4313444588C108583219 @default.
- W4313444588 hasConceptScore W4313444588C113775141 @default.
- W4313444588 hasConceptScore W4313444588C11413529 @default.
- W4313444588 hasConceptScore W4313444588C118524514 @default.
- W4313444588 hasConceptScore W4313444588C149635348 @default.
- W4313444588 hasConceptScore W4313444588C151927369 @default.
- W4313444588 hasConceptScore W4313444588C154945302 @default.
- W4313444588 hasConceptScore W4313444588C155032097 @default.
- W4313444588 hasConceptScore W4313444588C173608175 @default.
- W4313444588 hasConceptScore W4313444588C199360897 @default.
- W4313444588 hasConceptScore W4313444588C2777904410 @default.
- W4313444588 hasConceptScore W4313444588C2780513914 @default.
- W4313444588 hasConceptScore W4313444588C28855332 @default.
- W4313444588 hasConceptScore W4313444588C41008148 @default.
- W4313444588 hasConceptScore W4313444588C45374587 @default.
- W4313444588 hasConceptScore W4313444588C50644808 @default.
- W4313444588 hasConceptScore W4313444588C68339613 @default.
- W4313444588 hasConceptScore W4313444588C9390403 @default.
- W4313444588 hasLocation W43134445881 @default.
- W4313444588 hasOpenAccess W4313444588 @default.
- W4313444588 hasPrimaryLocation W43134445881 @default.
- W4313444588 hasRelatedWork W2055789895 @default.
- W4313444588 hasRelatedWork W2356033780 @default.
- W4313444588 hasRelatedWork W2898755250 @default.
- W4313444588 hasRelatedWork W2949493108 @default.
- W4313444588 hasRelatedWork W3092089946 @default.
- W4313444588 hasRelatedWork W3102452718 @default.
- W4313444588 hasRelatedWork W3216454443 @default.
- W4313444588 hasRelatedWork W4283718600 @default.
- W4313444588 hasRelatedWork W4313444588 @default.
- W4313444588 hasRelatedWork W4315473692 @default.
- W4313444588 isParatext "false" @default.
- W4313444588 isRetracted "false" @default.
- W4313444588 workType "article" @default.