arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1909.12291cs.LGcs.DCstat.ML
收藏

Exascale深度学习加速癌症研究

Exascale Deep Learning to Accelerate Cancer Research

Oak Ridge National Lab(橡树岭国家实验室);Stony Brook University(石溪大学)
Robert M. Patton, J. Travis Johnston, Steven R. Young, Catherine D. Schuman, Thomas E. Potok, Derek C. Rose, Seung-Hwan Lim, Junghoon Chae, Le Hou, Shahira Abou… 展开作者

Robert M. Patton, J. Travis Johnston, Steven R. Young, Catherine D. Schuman, Thomas E. Potok, Derek C. Rose, Seung-Hwan Lim, Junghoon Chae, Le Hou, Shahira Abousamra, Dimitris Samaras, Joel Saltz

摘要:

通过使用神经网络,深度学习已经证明了在提供足够的数据进行训练时自动化许多常规任务的卓越能力。神经网络架构(例如,层的数量、层的类型、层之间的连接等)在确定神经网络能够从训练数据学习什么(如果有的话)方面起着关键作用。神经网络架构的趋势,特别是那些在ImageNet上训练的架构,已经变得越来越深,越来越复杂。其结果是基准数据集的准确性不断提高,但代价是计算需求增加。在本文中,我们证明了神经网络架构可以自动生成,为特定的应用程序量身定制,具有双重目标:预测精度和预测速度。使用MENNDL--一个支持HPC的神经架构搜索软件栈--我们在癌症病理数据集上生成了一个神经网络,其准确性与最先进的网络相当,而且推理速度也比$16\times$快。由于癌症病理学数据的数量和速度,推理的加速是必要的;具体来说,对于没有访问HPC系统的个人研究人员来说,以前最先进的网络太慢,无法跟上数据生成的速度。我们的新模型使具有适度计算资源的研究人员能够比收集更快地分析新生成的数据。

英文摘要:

Deep learning, through the use of neural networks, has demonstrated remarkable ability to automate many routine tasks when presented with sufficient data for training. The neural network architecture (e.g. number of layers, types of layers, connections between layers, etc.) plays a critical role in determining what, if anything, the neural network is able to learn from the training data. The trend for neural network architectures, especially those trained on ImageNet, has been to grow ever deeper and more complex. The result has been ever increasing accuracy on benchmark datasets with the cost of increased computational demands. In this paper we demonstrate that neural network architectures can be automatically generated, tailored for a specific application, with dual objectives: accuracy of prediction and speed of prediction. Using MENNDL--an HPC-enabled software stack for neural architecture search--we generate a neural network with comparable accuracy to state-of-the-art networks on a cancer pathology dataset that is also $16\times$ faster at inference. The speedup in inference is necessary because of the volume and velocity of cancer pathology data; specifically, the previous state-of-the-art networks are too slow for individual researchers without access to HPC systems to keep pace with the rate of data generation. Our new model enables researchers with modest computational resources to analyze newly generated data faster than it is collected.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑