2,497 publications from this institution
The use of increasingly larger and more complex neural networks (NNs) makes it critical to scale the capabilities and efficiency of NN accelerators. Tiled architectures provide an intuitive scaling solution that supports both coarse-grained parallelism in NNs: intra-layer parallelism, where all tiles process a single layer, and inter-layer pipelining, where multiple layers execute across tiles in a pipelined manner. This work proposes dataflow optimizations to address the shortcomings of existing parallel dataflow techniques for tiled NN accelerators. For intra-layer parallelism, we develop buffer sharing dataflow that turns the distributed buffers into an idealized shared buffer, eliminating excessive data duplication and the memory access overheads. For inter-layer pipelining, we develop alternate layer loop ordering that forwards the intermediate data in a more fine-grained and timely manner, reducing the buffer requirements and pipeline delays. We also make inter-layer pipelining applicable to NNs with complex DAG structures. These optimizations improve the performance of tiled NN accelerators by 2x and reduce their energy consumption by 45% across a wide range of NNs. The effectiveness of our optimizations also increases with the NN size and complexity.
Meta-analysis is an important research design for appraising evidence and guiding medical practice and health policy. [1][1] Meta-analyses draw strength from combining data from many studies. However, even if perfectly done with perfect data, a single meta-analysis that addresses 1 treatment
Most of the RCD studies we identified explored comparative treatment effects that had already been investigated in RCTs. The objective of such studies needs to shift more toward answering pivotal questions that are not supported by trial evidence or for which RCTs would be unfeasible.
OPINION article Front. Genet., 14 March 2013Sec. Statistical Genetics and Methodology Volume 4 - 2013 | https://doi.org/10.3389/fgene.2013.00033