Xiaobing Feng

TL;DR: This paper proposes an accelerator which is 60x more energy efficient than the previous state-of-the-art neural network accelerator, designed down to the layout at 65 nm, with a modest footprint and consuming only 320 mW, but still about 30x faster than high-end GPUs.

TL;DR: An ML accelerator called PuDianNao is presented, which accommodates seven representative ML techniques, including k-means, k-nearest neighbors, naive bayes, support vector machine, linear regression, classification tree, and deep neural network, and can perform up to 1056 GOP/s, and consumes 596 mW only.

TL;DR: The level-by-level algorithm, LevPA, gives rises to a precise and compact SSA representation for subsequent program analysis and optimization tasks and a flow- and context-sensitive MAY/MUST mod (modification) set and read set for each procedure.

TL;DR: A new hardware and software cooperative DRAM bank partitioning method that combines page coloring and XOR cache mapping to evaluate the benefit potential of reducing interthread interference is presented.

TL;DR: Panthera is proposed, a semantics-aware, fully automated memory management technique for Big Data processing over hybrid memories that reduces energy by 32 – 52% at only a 1 – 9% execution time overhead.