ORNL researchers design novel method for energy-efficient deep neural networks

March 14, 2018

An Oak Ridge National Laboratory method to improve the energy efficiency of scientific artificial intelligence is showing early promise in efforts to parse insights from volumes of cancer data.

Researchers are realizing the potential of deep learning to rapidly advance science, but "training" the underlying neural networks with large volumes of data to tackle the task at hand can require large amounts of energy. These networks also require complex connectivity and enormous amounts of storage, both of which further reduce their energy efficiency and potential in real-world applications.

To address this issue, ORNL's Mohammed Alawad, Hong-Jun Yoon, and Georgia Tourassi developed a novel method for the development of energy-efficient deep neural networks capable of solving complex science problems. They presented their research at the 2017 IEEE Conference on Big Data in Boston.

The researchers demonstrated that by converting deep learning neural networks (DNNs) to "deep spiking" neural networks (DSNNs), they can improve the energy efficiency of network design and realization.

DSNNs imitate neurons in the human brain via pulses or "spikes" in the place of actual signals, with the individual spikes indicating where to perform the computations. This process minimizes the necessary calculations and maximizes the network's energy efficiency. However, energy efficiency comes at the cost of task performance, and the authors' novel stochastic method for implementing DSNNs overcomes this tradeoff.

The results were impressive: the team's approach achieved nearly the same accuracy as the original DNN and performed better than a state-of-the-art spiking neural network. The team's stochastic-based DSNN, which distributes spikes uniformly over time, consumed 38 times less energy than the original DNN and almost 2 times less energy than a conventional DSNN while delivering markedly better task performance.

The researchers trained their network on clinical text data from the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) Program, which provides cancer statistics such as incidence, prevalence, and mortality across the population associated by age, sex, race, year of diagnosis, and geographic areas.

The ORNL team applied the newly trained networks to clinical pathology reports, the main source of information for the national cancer surveillance program. These reports contain vast amounts of unstructured text, said Yoon, and researchers are developing intelligent language understanding systems to extract the most relevant clinical concepts in the sea of text.

The clinical reports represent a "sparse" dataset, which typically pose unique challenges to spiking networks. Most DSNN techniques have focused on computer vision tasks such as the MNIST dataset, which consist of a series of handwritten digits to train image processing networks. These datasets are typically "dense," meaning all variables in the dataset are populated with values, a characteristic that often simplifies analyses.

Traditional techniques for improving the performance and energy efficiency of spiking networks often preserve the structures of conventional neural networks, a practice that sacrifices accuracy and performance. These shortcomings drove the team to develop a new methodology, which relies on simple circuit hardware to perform complex calculations.

"Spiking the network lowers energy consumption because we disregard the unnecessary computations and we look only for the relevant nodes of the network," said Yoon, "and this is one way we get energy efficiency improvements while identifying important clinical information with high accuracy."

The team's technique will help ORNL researchers in the CANcer Distributed Learning Environment (CANDLE) project, which aims to use the lab's world-class big data expertise and computing facilities to scan millions of clinical reports in search for insights on causes of cancer, best courses of treatment, and improved outcomes. They will soon attempt to parallelize the algorithm for more computational efficiency.

The spiking networks were optimized on graphics processing units (GPUs), the processors of choice for artificial intelligent applications, particularly those that utilize machine learning and deep learning. However, the methodology can be extended for training spiking networks, further increasing the energy efficiency of these novel networks while expediting scientific discovery via deep learning.
-end-
The research was supported by ORNL's Laboratory Directed Research and Development program and the Joint Design of Advanced Computing Solutions for Cancer program established by DOE and the NCI.

ORNL is managed by UT-Battelle for DOE's Office of Science, the single largest supporter of basic research in the physical sciences in the United States. DOE's Office of Science is working to address some of the most pressing challenges of our time. For more information, please visit http://science.energy.gov/.

DOE/Oak Ridge National Laboratory

Related Cancer Articles from Brightsurf:

New blood cancer treatment works by selectively interfering with cancer cell signalling
University of Alberta scientists have identified the mechanism of action behind a new type of precision cancer drug for blood cancers that is set for human trials, according to research published in Nature Communications.

UCI researchers uncover cancer cell vulnerabilities; may lead to better cancer therapies
A new University of California, Irvine-led study reveals a protein responsible for genetic changes resulting in a variety of cancers, may also be the key to more effective, targeted cancer therapy.

Breast cancer treatment costs highest among young women with metastic cancer
In a fight for their lives, young women, age 18-44, spend double the amount of older women to survive metastatic breast cancer, according to a large statewide study by the University of North Carolina at Chapel Hill.

Cancer mortality continues steady decline, driven by progress against lung cancer
The cancer death rate declined by 29% from 1991 to 2017, including a 2.2% drop from 2016 to 2017, the largest single-year drop in cancer mortality ever reported.

Stress in cervical cancer patients associated with higher risk of cancer-specific mortality
Psychological stress was associated with a higher risk of cancer-specific mortality in women diagnosed with cervical cancer.

Cancer-sniffing dogs 97% accurate in identifying lung cancer, according to study in JAOA
The next step will be to further fractionate the samples based on chemical and physical properties, presenting them back to the dogs until the specific biomarkers for each cancer are identified.

Moffitt Cancer Center researchers identify one way T cell function may fail in cancer
Moffitt Cancer Center researchers have discovered a mechanism by which one type of immune cell, CD8+ T cells, can become dysfunctional, impeding its ability to seek and kill cancer cells.

More cancer survivors, fewer cancer specialists point to challenge in meeting care needs
An aging population, a growing number of cancer survivors, and a projected shortage of cancer care providers will result in a challenge in delivering the care for cancer survivors in the United States if systemic changes are not made.

New cancer vaccine platform a potential tool for efficacious targeted cancer therapy
Researchers at the University of Helsinki have discovered a solution in the form of a cancer vaccine platform for improving the efficacy of oncolytic viruses used in cancer treatment.

American Cancer Society outlines blueprint for cancer control in the 21st century
The American Cancer Society is outlining its vision for cancer control in the decades ahead in a series of articles that forms the basis of a national cancer control plan.

Read More: Cancer News and Cancer Current Events
Brightsurf.com is a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to Amazon.com.