Intel announces Fully-fused Multi-Layer Perceptrons on Intel Data Center GPUs https://huggingface.co/papers/2403.
This paper presents a SYCL implementation of Multi-Layer Perceptrons (MLPs), which targets and is optimized for the Intel Data Center GPU Max 1550.
Join the discussion on this paper page.
Implementing fully-fused Multi-Layer Perceptrons (MLPs) on Intel Data Center GPUs marks a significant advancement in leveraging hardware capabilities for deep learning tasks. This integration optimizes performance and efficiency, streamlining complex computations inherent in MLP architectures.