Deep Learning

A Performance Analysis Framework for Exploiting GPU Microarchitectural Capability

Presented our ICS'17 work.

Deep Learning on Modern Architectures

Discussed how state-of-the-art deep learning libraries optimize computations by utilizing architectural features.

Understanding the GPU microarchitecture to achieve bare-metal performance tuning

Convolution Methods

Introduced various kinds of convolution methods and analyzed their complexities, memory consumptions, and data access patterns.