Dimensions overall score 7.0

GitHub Code Pulse

No public code linked for this paper yet.

• The proposed Optical In-Network-Computing (OptINC) architecture can eliminate communication overhead in distributed learning compared to the ring all-reduce baseline.result 90%
• OptINC uses Mach-Zehnder-Interferometers (MZIs) to implement gradient averaging and quantization computations directly in the optical domain.method 95%
• A preprocessing unit implemented in the optical domain reduces the dataset complexity for training the optical neural network (ONN) used in OptINC.method 85%
• Hardware cost of the optical neural network is lowered by approximating its weight matrices with unitary and diagonal matrices.technical 90%
• A hardware-aware training algorithm is proposed to maintain the accuracy of the ONN despite the hardware approximations.method 90%
• Optical circuit switches (OCS) are suitable for distributed learning because communication patterns are predetermined, making their reconfiguration latency negligible.technical 85%
• An arbitrary M×N matrix can be implemented with MZIs by decomposing it via Singular Value Decomposition (SVD) into unitary and diagonal components.technical 95%
• The proposed OptINC framework achieves comparable training accuracy to the ring all-reduce baseline on tasks like ResNet50 on CIFAR-100 and a LLaMA-based network on Wikipedia-1B.result 90%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card