Abstract-- Benefitting from large-scale training datasets and the complex training network, Convolutional Neural Networks (CNNs) are widely applied in various fields with high accuracy. However, the training process of CNNs is very time-consuming, where large amounts of training samples and iterative operations are required to obtain high-quality weight parameters. In this paper, we focus on the time-consuming training process of large-scale CNNs and propose a Bi-layered Parallel Training (BPT-CNN) architecture in distributed computing environments. BPT-CNN consists of two main components: (a) an outer-layer parallel training for multiple CNN subnetworks on separate data subsets, and (b) an inner-layer parallel training for each subnetwork. In the outer-layer parallelism, we address critical issues of distributed and parallel computing, including data communication, synchronization, and workload balance. A heterogeneousaware Incremental Data Partitioning and Allocation (IDPA) strategy is proposed, where large-scale training datasets are partitioned and allocated to the computing nodes in batches according to their computing power. To minimize the synchronization waiting during the global weight update process, an Asynchronous Global Weight Update (AGWU) strategy is proposed. In the inner-layer parallelism, we further accelerate the training process for each CNN subnetwork on each computer, where computation steps of convolutional layer and the local weight training are parallelized based on task-parallelism. We introduce task decomposition and scheduling strategies with the objectives of thread-level load balancing and minimum waiting time for critical paths. Extensive experimental results indicate that the proposed BPT-CNN effectively improves the training performance of CNNs while maintaining the accuracy. Index Terms--Big data, bi-layered parallel computing, convolutional neural networks, deep learning, distributed computing. Convolutional Neural Network (CNN) algorithm is an important branch of DL.

Protein interactions constitute the fundamental building block of almost every life activity. Identifying protein communities from Protein-Protein Interaction (PPI) networks is essential to understand the principles of cellular organization and explore the causes of various diseases. It is critical to integrate multiple data resources to identify reliable protein communities that have biological significance and improve the performance of community detection methods for large-scale PPI networks. In this paper, we propose a Multi-source Learning based Protein Community Detection (MLPCD) algorithm by integrating Gene Expression Data (GED) and a parallel solution of MLPCD using cloud computing technology. To effectively discover the biological functions of proteins that participating in different cellular processes, GED under different conditions is integrated with the original PPI network to reconstruct a Weighted-PPI (WPPI) network. To flexibly identify protein communities of different scales, we define community modularity and functional cohesion measurements and detect protein communities from WPPI using an agglomerative method. In addition, we respectively compare the detected communities with known protein complexes and evaluate the functional enrichment of protein function modules using Gene Ontology annotations. Moreover, we implement a parallel version of the MLPCD algorithm on the Apache Spark platform to enhance the performance of the algorithm for large-scale realistic PPI networks. Extensive experimental results indicate the superiority and notable advantages of the MLPCD algorithm over the relevant algorithms in terms of accuracy and performance.

Buzzwords like "deep learning" and "neural networks" are everywhere, but so much of the popular understanding is misguided, says Terrence Sejnowski, a computational neuroscientist at the Salk Institute for Biological Studies. Sejnowski, a pioneer in the study of learning algorithms, is the author of The Deep Learning Revolution (out next week from MIT Press). He argues that the hype about killer AI or robots making us obsolete ignores exciting possibilities happening in the fields of computer science and neuroscience, and what can happen when artificial intelligence meets human intelligence. The Verge spoke to Sejnkowski about how "deep learning" suddenly became everywhere, what it can and cannot do, and the problem of hype. This interview has been lightly edited for clarity.

Personal computers, conservation, pro football, rock n' roll and rocket ships: Paul G. Allen couldn't have asked for a better way to spend, invest and donate the billions he reaped from co-founding Microsoft with childhood friend Bill Gates. Allen used the fortune he made from Microsoft -- whose Windows operating system is found on most of the world's desktop computers -- to invest in other ambitions, from tackling climate change and advancing brain research to finding innovative solutions to solve some of the world's biggest challenges. "If it has the potential to do good, then we should do it," Gates quoted his friend as saying. Allen died Monday in Seattle from complications of non-Hodgkin's lymphoma, according to his company Vulcan Inc. Just two weeks ago, Allen, who owned the NFL's Seattle Seahawks and the NBA's Portland Trail Blazers, had announced that the same cancer he had in 2009 had returned.

In the spring of 2011, a small group of engineers working on a secretive project at Google received an e-mail from a colleague. Anthony is going to get fired. Several of the recipients gathered in one of the self-serve espresso bars that dot the company's headquarters, and traded rumors suggesting that Anthony Levandowski--one of the company's most talented and best-known employees--had finally gone too far. Levandowski was a gifted engineer who frequently spoke to newspapers and magazines, including this one, about the future of robotics. On the Google campus, he was easy to pick out: he was six feet seven and wore the same drab clothes every day--jeans and a gray T-shirt--which, in Silicon Valley, signalled that he preferred to conserve his cognitive energies for loftier pursuits.

Billionaire Paul Allen, who founded US software giant Microsoft with Bill Gates in the 1970s, died on Monday at the age of 65 after his latest battle with cancer, his family said. Allen said earlier this month he was being treated for non-Hodgkin's lymphoma, the same kind of cancer he battled and overcame nearly a decade ago. He was first diagnosed when he was CEO of Microsoft. Allen was the man who persuaded school-friend Bill Gates to drop out of Harvard to start what became the world's biggest software company. Allen left Microsoft in 1983 - before the company became a corporate juggernaut - following a dispute with Gates, but his share of their original partnership allowed him to spend the rest of his life and billions of dollars on yachts, art, rock music, sports teams, brain research and real estate.

A link has been posted to your Facebook feed. Paul Allen, sports owner, 1953-2018 (Photo: Ted S. Warren, AP) SAN FRANCISCO – Paul Allen, a technology pioneer who helped launch the personal computer revolution as co-founder of Microsoft with Bill Gates, has died, according to his company, Vulcan Inc. The cause was complications from non-Hodgkin's lymphoma, a condition that surfaced in 2009 and returned just a few weeks ago. On Oct. 1, Allen wrote a short but optimistic note on his personal website, noting that "I've begun treatment & my doctors are optimistic that I will see a good result. Appreciate the support I've received & count on it as I fight this challenge."

We propose two sparsity-aware normalized subband adaptive filter (NSAF) algorithms by using the gradient descent method to minimize a combination of the original NSAF cost function and the l1-norm penalty function on the filter coefficients. This l1-norm penalty exploits the sparsity of a system in the coefficients update formulation, thus improving the performance when identifying sparse systems. Compared with prior work, the proposed algorithms have lower computational complexity with comparable performance. We study and devise statistical models for these sparsity-aware NSAF algorithms in the mean square sense involving their transient and steady -state behaviors. This study relies on the vectorization argument and the paraunitary assumption imposed on the analysis filter banks, and thus does not restrict the input signal to being Gaussian or having another distribution. In addition, we propose to adjust adaptively the intensity parameter of the sparsity attraction term. Finally, simulation results in sparse system identification demonstrate the effectiveness of our theoretical results.

I feel that there was a sort of explosion a couple of years ago after which the whole topic of Artificial Intelligence (AI) suddenly sprang into a wider audience's consciousness. All of a sudden we had Siri, Amazon's Alexa and we started talking about self-driving cars. Jaan Tallinn, how did it happen? There were two different explosions. I believe that a lot of the latter had to do with the works of Elon Musk and Stephen Hawking. Most importantly, the former was the revolution of deep learning.

Artificial intelligence is shaking-up the legal industry. This includes some breakthrough AI technology from Thomson Reuters, which will affect how attorneys work with their clients. Examples include changing billable hours, competitive advantages to building cases, and so on. To find out more about the new AI technology and to understand how new technologies are disrupting law in general, Digital Journal caught up with Dr. Khalid Al-Kofahi from Thomson Reuters. Digital Journal: How has artificial intelligence advanced in the legal space in recent years?