Conference proceeding
Sparse Compressed Quantized Dataflow Architecture (QUDA) for Neuromorphic Computing
Proceedings ... Annual IEEE Symposium on Field-Programmable Custom Computing Machines (Online), pp 283-283
13 May 2026
Abstract
Compressed Sparse Row (CSR) encoding stores only non-zero elements of a sparse matrix along with indexing information for efficient MVM. We propose QUDA, a heterogeneous 2D dataflow architecture that operates directly on CSR-encoded, quantized weights to improve memory efficiency and utilization. It combines flow-through systolic array (FTSA) columns for partial accumulation with a final output-stationary systolic array (OSSA) column for accumulation and spike generation. Our results show significant improvements in resource utilization and power consumption compared to state-of-the-art designs.
Metrics
1 Record Views
Details
- Title
- Sparse Compressed Quantized Dataflow Architecture (QUDA) for Neuromorphic Computing
- Creators
- Shadi Matinizadeh - Drexel UniversityAnup Das - Drexel University
- Publication Details
- Proceedings ... Annual IEEE Symposium on Field-Programmable Custom Computing Machines (Online), pp 283-283
- Publisher
- IEEE
- Resource Type
- Conference proceeding
- Language
- English
- Academic Unit
- Electrical and Computer Engineering; College of Engineering
- Other Identifier
- 991022191263204721