From the 1 of 4 linked papers with an AI index.
4 papers
NIFA: Nonlinear IMC enhanced FPGA for efficient ML inference
Jiajun Hu, Ruthwik Reddy Sunketa, Lei Zhao +3
The paper proposes a new FPGA architecture that replaces ADCs with analog content‑addressable memories to enable ADC‑free in‑memory computing, allowing both linear and nonlinear op…
ATLAS: Automated HLS for DL-Optimized FPGAs
Ruthwik Reddy Sunketa, Aman Arora
FPGA architectures increasingly incorporate domain-specific in-fabric hardblocks to accelerate DL inference, particularly GEMM, which dominates DL computation. To realize the perfo…
Boosting FPGA Performance with Direct BRAM-DSP Paths
Jiajun Hu, Ruthwik Reddy Sunketa, Andrew Boutros +1
Efficient data movement between memory and compute units is a key performance bottleneck in modern FPGA designs, particularly for deep learning (DL) workloads. In typical FPGA arch…
Programming Domain-Specific FPGA Hardblocks from HLS: An RTL Blackbox Approach
Ruthwik Reddy Sunketa, Jeevesh Choudhury, Aman Arora
Domain-specific Field Programmable Gate Array (FPGA) architectures increasingly integrate specialized hardblocks, such as Tensor Slices, to accelerate artificial intelligence and m…