Searcharxiv⌕ Search

arXiv subjects

Mark Horowitz

Publications and source records attributed to Mark Horowitz.

24 records · Page 2Linked to original sources
arXiv2018

Training Domain Specific Models for Energy-Efficient Object Detection

We propose an end-to-end framework for training domain specific models (DSMs) to obtain both high accuracy and computational efficiency for object detection tasks. DSMs are trained with distillation \cite{hinton2015distilling} and focus on achieving high accuracy at a limited domain (e.g. fixed view of an intersection). We argue that DSMs can capture essential features well even with a small model size, enabling higher accuracy and efficiency than traditional techniques. In addition, we improve the training efficiency by reducing the dataset size by culling easy to classify images from the training set. For the limited domain, we observed that compact DSMs significantly surpass the accuracy of COCO trained models of the same size. By training on a compact dataset, we show that with an accuracy drop of only 3.6\%, the training time can be reduced by 93\%. The codes are uploaded in https://github.com/kentaroy47/training-domain-specific-models.

cs.CV↗
arXiv2016

Programming Heterogeneous Systems from an Image Processing DSL

Specialized image processing accelerators are necessary to deliver the performance and energy efficiency required by important applications in computer vision, computational photography, and augmented reality. But creating, "programming,"and integrating this hardware into a hardware/software system is difficult. We address this problem by extending the image processing language, Halide, so users can specify which portions of their applications should become hardware accelerators, and then we provide a compiler that uses this code to automatically create the accelerator along with the "glue" code needed for the user's application to access this hardware. Starting with Halide not only provides a very high-level functional description of the hardware, but also allows our compiler to generate the complete software program including the sequential part of the workload, which accesses the hardware for acceleration. Our system also provides high-level semantics to explore different mappings of applications to a heterogeneous system, with the added flexibility of being able to map at various throughput rates. We demonstrate our approach by mapping applications to a Xilinx Zynq system. Using its FPGA with two low-power ARM cores, our design achieves up to 6x higher performance and 8x lower energy compared to the quad-core ARM CPU on an NVIDIA Tegra K1, and 3.5x higher performance with 12x lower energy compared to the K1's 192-core GPU.

cs.SE↗
arXiv2016

FPMax: a 106GFLOPS/W at 217GFLOPS/mm2 Single-Precision FPU, and a 43.7GFLOPS/W at 74.6GFLOPS/mm2 Double-Precision FPU, in 28nm UTBB FDSOI

FPMax implements four FPUs optimized for latency or throughput workloads in two precisions, fabricated in 28nm UTBB FDSOI. Each unit's parameters, e.g pipeline stages, booth encoding etc., were optimized to yield 1.42ns latency at 110GLOPS/W (SP) and 1.39ns latency at 36GFLOPS/W (DP). At 100% activity, body-bias control improves the energy efficiency by about 20%; at 10% activity this saving is almost 2x. Keywords: FPU, energy efficiency, hardware generator, SOI

cs.AR↗
arXiv2016

A Systematic Approach to Blocking Convolutional Neural Networks

Convolutional Neural Networks (CNNs) are the state of the art solution for many computer vision problems, and many researchers have explored optimized implementations. Most implementations heuristically block the computation to deal with the large data sizes and high data reuse of CNNs. This paper explores how to block CNN computations for memory locality by creating an analytical model for CNN-like loop nests. Using this model we automatically derive optimized blockings for common networks that improve the energy efficiency of custom hardware implementations by up to an order of magnitude. Compared to traditional CNN CPU implementations based on highly-tuned, hand-optimized BLAS libraries,our x86 programs implementing the optimal blocking reduce the number of memory accesses by up to 90%.

cs.DC↗
arXiv2009

Towards an explanatory and computational theory of scientific discovery

We propose an explanatory and computational theory of transformative discoveries in science. The theory is derived from a recurring theme found in a diverse range of scientific change, scientific discovery, and knowledge diffusion theories in philosophy of science, sociology of science, social network analysis, and information science. The theory extends the concept of structural holes from social networks to a broader range of associative networks found in science studies, especially including networks that reflect underlying intellectual structures such as co-citation networks and collaboration networks. The central premise is that connecting otherwise disparate patches of knowledge is a valuable mechanism of creative thinking in general and transformative scientific discovery in particular.

cs.GL↗