Toxicity analysis and prediction are of paramount importance to human health and environmental protection. Existing computational methods are built from a wide variety of descriptors and regressors, which makes their performance analysis difficult. For example, deep neural network (DNN), a successful approach in many occasions, acts like a black box and offers little conceptual elegance or physical understanding. The present work constructs a common set of microscopic descriptors based on established physical models for charges, surface areas and free energies to assess the performance of multi-task convolutional neural network (MT-CNN) architectures and a few other approaches, including random forest (RF) and gradient boosting decision tree (GBDT), on an equal footing. Comparison is also given to convolutional neural network (CNN) and non-convolutional deep neural network (DNN) algorithms. Four benchmark toxicity data sets (i.e., endpoints) are used to evaluate various approaches. Extensive numerical studies indicate that the present MT-CNN architecture is able to outperform the state-of-the-art methods.

Automatically detecting sound units of humpback whales in complex time-varying background noises is a current challenge for scientists. In this paper, we explore the applicability of Convolution Neural Network (CNN) method for this task. In the evaluation stage, we present 6 bi-class classification experimentations of whale sound detection against different background noise types (e.g., rain, wind). In comparison to classical FFT-based representation like spectrograms, we showed that the use of image-based pretrained CNN features brought higher performance to classify whale sounds and background noise.

Objective: In this work, we perform margin assessment of human breast tissue from optical coherence tomography (OCT) images using deep neural networks (DNNs). This work simulates an intraoperative setting for breast cancer lumpectomy. Methods: To train the DNNs, we use both the state-of-the-art methods (Weight Decay and DropOut) and a newly introduced regularization method based on function norms. Commonly used methods can fail when only a small database is available. The use of a function norm introduces a direct control over the complexity of the function with the aim of diminishing the risk of overfitting. Results: As neither the code nor the data of previous results are publicly available, the obtained results are compared with reported results in the literature for a conservative comparison. Moreover, our method is applied to locally collected data on several data configurations. The reported results are the average over the different trials. Conclusion: The experimental results show that the use of DNNs yields significantly better results than other techniques when evaluated in terms of sensitivity, specificity, F1 score, G-mean and Matthews correlation coefficient. Function norm regularization yielded higher and more robust results than competing methods. Significance: We have demonstrated a system that shows high promise for (partially) automated margin assessment of human breast tissue, Equal error rate (EER) is reduced from approximately 12\% (the lowest reported in the literature) to 5\%\,--\,a 58\% reduction. The method is computationally feasible for intraoperative application (less than 2 seconds per image).

Person recognition aims at recognizing the same identity across time and space with complicated scenes and similar appearance. In this paper, we propose a novel method to address this task by training a network to obtain robust and representative features. The intuition is that we directly compare and optimize the cosine distance between two features - enlarging inter-class distinction as well as alleviating inner-class variance. We propose a congenerous cosine loss by minimizing the cosine distance between samples and their cluster centroid in a cooperative way. Such a design reduces the complexity and could be implemented via softmax with normalized inputs. Our method also differs from previous work in person recognition that we do not conduct a second training on the test subset. The identity of a person is determined by measuring the similarity from several body regions in the reference set. Experimental results show that the proposed approach achieves better classification accuracy against previous state-of-the-arts.

Canadian researchers have been behind some recent major breakthroughs in artificial intelligence. Now, the country is betting on becoming a big player in one of the hottest fields in technology, with help from the likes of Google and RBC. In an unassuming building on the University of Toronto's downtown campus, Geoff Hinton laboured for years on the "lunatic fringe" of academia and artificial intelligence, pursuing research in an area of AI called neural networks. Also known as "deep learning", neural networks are computer programs that learn in a similar way to human brains. The field showed early promise in the 1980s, but the tech sector turned its attention to other AI methods after that promise seemed slow to develop.

Ontario is ensuring businesses in the province continue to stay ahead in the innovation economy, and attract investment and top talent, by supporting a new institute for artificial intelligence (AI) with a specific focus on deep learning and machine learning. Ontario's highly skilled workforce, competitive business climate and recognized culture of innovation make the province an ideal home for the Vector Institute for artificial intelligence. The institute will work to produce, retain and attract top talent in the field and generate further investment from companies looking to hire experts and expand their AI divisions. Collaborating with established businesses, start-ups and academic institutions, the institute's work holds great promise for solving some of the biggest health, social and economic challenges of our time. Artificial intelligence can deliver improvements as diverse as accurately diagnosing an illness earlier, to protecting people against consumer fraud. Ontario's investment in the Vector Institute will help: Staying on the cutting edge of technology and leading global innovation is part of Ontario's plan to create jobs, grow our economy and help people in their everyday lives.

In 1971, Terry Winograd wrote the SHRDLU program while completing his PhD at MIT. SHRDLU features a world of toy blocks where the computer translates human commands into physical actions, such as "move the red pyramid next to the blue cube." To succeed in such tasks, the computer must build up semantic knowledge iteratively, a process Winograd discovered was brittle and limited. The rise of chatbots and voice activated technologies has renewed fervor in natural language processing (NLP) and natural language understanding (NLU) techniques that can produce satisfying human-computer dialogs. Unfortunately, academic breakthroughs have not yet translated to improved user experiences, with Gizmodo writer Darren Orf declaring Messenger chatbots "frustrating and useless" and Facebook admitting a 70% failure rate for their highly anticipated conversational assistant M. Nevertheless, researchers forge ahead with new plans of attack, occasionally revisiting the same tactics and principles Winograd tried in the 70s. OpenAI recently leveraged reinforcement learning to teach to agents to design their own language by "dropping them into a set of simple worlds, giving them the ability to communicate, and then giving them goals that can be best achieved by communicating with other agents."

Google and RBC and others are helping Canada shine in the artificial intelligence field, BBC reports. Toronto will soon get the Vector Institute for Artificial Intelligence, aimed at fuelling "Canada's amazing AI momentum". The new research facility will be officially launched on Tuesday, and what makes it special is that it will be dedicated to expanding the applications of AI through explorations in deep learning and other forms of machine learning. The institute received CAD$170 million in funding from the Canadian and Ontario governments, and a group of businesses that includes both Google and RBC. Geoff Hinton, considered the godfather of deep learning, will be the research institute's chief scientific adviser.

This webinar focuses on demonstrating how the Data Science Virtual Machine (DSVM) in Microsoft Azure conveniently enables key end-to-end data analytics scenarios by providing users immediate access to a collection of the top data science and development tools of the industry, completely pre-configured, with worked out examples and sample code. We will do a detailed demonstration of some key capabilities of the DSVM by working through a selection of popular scenarios using technologies that are enabled by it. These examples encompass areas such as using a local Spark environment for easy test and development, training and scoring for deep-learning on GPU based instances of the DSVM, cross-platform data exploration and querying using Apache Drill, and in-database analytics using SQL Server 2016 R Services. Both the Windows and Linux flavors of the VMs are covered in this webinar.

Many companies today are employing deep learning techniques in different facets of their business. Yelp uses deep learning algorithms to feature the best user photos, Netflix uses it to suggest movies you might be interested in, and Google ultimately transformed the concept of deep learning by creating a system that helps generate responses to search queries. It's widely believed that you no longer need to know your data; you can just apply a little deep learning magic and poof -- problem solved. However, the reality is that this could not be further from the truth, at least for the legal industry. As an example, deep learning can be essential when legal counsel within an organization wants to find out how many contracts (among 10s to 100s of thousands) has termination for convenience clauses that could disrupt the business, or if any have strict assignment rules that may be a problem for a M&A event.