Fluorescence microscopy has enabled a dramatic development in modern biology. Due to its inherently weak signal, fluorescence microscopy is not only much noisier than photography, but also presented with Poisson-Gaussian noise where Poisson noise, or shot noise, is the dominating noise source, instead of Gaussian noise that dominates in photography. To get clean fluorescence microscopy images, it is highly desirable to have effective denoising algorithms and datasets that are specifically designed to denoise fluorescence microscopy images. While such algorithms exist, there are no such datasets available. In this paper, we fill this gap by constructing a dataset - the Fluorescence Microscopy Denoising (FMD) dataset - that is dedicated to Poisson-Gaussian denoising. The dataset consists 12,000 real fluorescence microscopy images obtained with commercial confocal, two-photon, and wide-field microscopes and representative biological samples such as cells, zebrafish, and mouse brain tissues. We use imaging averaging to effectively obtain ground truth images and 60,000 noisy images with different noise levels. We use this dataset to benchmark 10 representative denoising algorithms and find that deep learning methods have the best performance. To our knowledge, this is the first microscopy image dataset for Poisson-Gaussian denoising purposes and it could be an important tool for high-quality, real-time denoising applications in biomedical research.

Parameter reduction has been an important topic in deep learning due to the ever-increasing size of deep neural network models and the need to train and run them on resource limited machines. Despite many efforts in this area, there were no rigorous theoretical guarantees on why existing neural net compression methods should work. In this paper, we provide provable guarantees on some hashing-based parameter reduction methods in neural nets. First, we introduce a neural net compression scheme based on random linear sketching (which is usually implemented efficiently via hashing), and show that the sketched (smaller) network is able to approximate the original network on all input data coming from any smooth and well-conditioned low-dimensional manifold. The sketched network can also be trained directly via back-propagation. Next, we study the previously proposed HashedNets architecture and show that the optimization landscape of one-hidden-layer HashedNets has a local strong convexity property similar to a normal fully connected neural network. We complement our theoretical results with empirical verifications.

In this paper, a new deep reinforcement learning based augmented general sequence tagging system is proposed. The new system contains two parts: a deep neural network (DNN) based sequence tagging model and a deep reinforcement learning (DRL) based augmented tagger. The augmented tagger helps improve system performance by modeling the data with minority tags. The new system is evaluated on SLU and NLU sequence tagging tasks using ATIS and CoNLL-2003 benchmark datasets, to demonstrate the new system's outstanding performance on general tagging tasks. Evaluated by F1 scores, it shows that the new system outperforms the current state-of-the-art model on ATIS dataset by 1.9 % and that on CoNLL-2003 dataset by 1.4 %.

Recently, there has been great interest in the field of audio style transfer, where a stylized audio is generated by imposing the style of a reference audio on the content of a target audio. We improve on the current approaches which use neural networks to extract the content and the style of the audio signal and propose a new autoencoder based architecture for the task. This network generates a stylized audio for a content audio in a single forward pass. The proposed network architecture proves to be advantageous over the quality of audio produced and the time taken to train the network. The network is experimented on speech signals to confirm the validity of our proposal.

Graph Neural Networks (GNNs) for representation learning of graphs broadly follow a neighborhood aggregation framework, where the representation vector of a node is computed by recursively aggregating and transforming feature vectors of its neighboring nodes. Many GNN variants have been proposed and have achieved state-of-the-art results on both node and graph classification tasks. However, despite GNNs revolutionizing graph representation learning, there is limited understanding of their representational properties and limitations. Here, we present a theoretical framework for analyzing the expressive power of GNNs in capturing different graph structures. Our results characterize the discriminative power of popular GNN variants, such as Graph Convolutional Networks and GraphSAGE, and show that they cannot learn to distinguish certain simple graph structures. We then develop a simple architecture that is provably the most expressive among the class of GNNs and is as powerful as the Weisfeiler-Lehman graph isomorphism test. We empirically validate our theoretical findings on a number of graph classification benchmarks, and demonstrate that our model achieves state-of-the-art performance.

Android is the predominant mobile operating system for the past few years. The prevalence of devices that can be powered by Android magnetized not merely application developers but also malware developers with criminal intention to design and spread malicious applications that can affect the normal work of Android phones and tablets, steal personal information and credential data, or even worse lock the phone and ask for ransom. Researchers persistently devise countermeasures strategies to fight back malware. One of these strategies applied in the past five years is the use of deep learning methods in Android malware detection. This necessitates a review to inspect the accomplished work in order to know where the endeavors have been established, identify unresolved problems, and motivate future research directions. In this work, an extensive survey of static analysis, dynamic analysis, and hybrid analysis that utilized deep learning methods are reviewed with an elaborated discussion on their key concepts, contributions, and limitations.

Big data and AI are joined at the hip: the best AI applications require massive amounts of constantly updated training data to build state-of-the-art models. Increasingly more Spark users want to integrate Spark with distributed machine learning frameworks built for state-of-the-art training. Here's the problem: big data frameworks like Spark and distributed deep learning frameworks don't play well together due to the disparity between how big data jobs are executed and how deep learning jobs are executed. In this latest Data Science Central webinar, we'll share how Project Hydrogen, a Spark Project Improvement Proposal led by Databricks, is positioned as a potential solution to this dilemma.

Although deep learning holds enormous promise for advancing new discoveries in genomics, it also should be implemented mindfully and with appropriate caution. Deep learning should be applied to biological datasets of sufficient size, usually on the order of thousands of samples. The'black box' nature of deep neural networks is an intrinsic property and does not necessarily lend itself well to complete understanding or transparency. Subtle variations in the input data can have outsized effects and must be controlled for as well as possible. Importantly, deep learning methods should be compared with simpler machine learning models with fewer parameters to ensure that the additional model complexity afforded by deep learning has not led to overfitting of the data.

The role of the machine learning engineer has changed. In the past, a machine learning engineer was a software engineer with some knowledge of machine learning concepts. Today, a machine machine learning engineer is a software engineer who not only understands the latest machine learning and deep learning concepts but is able to deploy an AI system in production that is highly reliable, fast and scalable. In this course, you'll learn how to scale up your application and deploy them into production. By the end of the course, you'll have designed an ML/DL system, built a prototype and deployed a running application that can be accessed via API or web service.

Stephen Hawking has said, "The development of full artificial intelligence could spell the end of the human race." Elon Musk claims that AI is humanity's "biggest existential threat." That might have people asking: Wait, what? But these grand worries are rooted in research. Along with Hawking and Musk, prominent figures at Oxford and UC Berkeley and many of the researchers working in AI today believe that advanced AI systems, if deployed carelessly, could end all life on earth. This concern has been raised since the dawn of computing. But it has come into particular focus in recent years, as advances in machine-learning techniques have given us a more concrete understanding of what we can do with AI, what AI can do for (and to) us, and how much we still don't know. Some of them think advanced AI is so distant that there's no point in thinking about it now. Others are worried that excessive hype about the power of their field might kill it prematurely. And even among the people who broadly agree that AI poses unique dangers, there are varying takes on what steps make the most sense today.