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 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.

Developing new drugs is notoriously inefficient. Fewer than 12 percent of all drugs entering clinical trials end up in pharmacies, and it costs about $2.6 billion to bring a drug to market. There are so many molecules to test that pharmaceutical researchers use pipetting robots to test a few thousand variants all at once. The best candidates then go into animal models or cell cultures, where hopefully a few will go on to bigger animal and human clinical trials. Which is why more and more drug developers are turning to computers and artificial intelligence to narrow down the list of potential drug molecules--saving time and money on those downstream tests.

We consider the problem of low canonical polyadic (CP) rank tensor completion. A completion is a tensor whose entries agree with the observed entries and its rank matches the given CP rank. We analyze the manifold structure corresponding to the tensors with the given rank and define a set of polynomials based on the sampling pattern and CP decomposition. Then, we show that finite completability of the sampled tensor is equivalent to having a certain number of algebraically independent polynomials among the defined polynomials. Our proposed approach results in characterizing the maximum number of algebraically independent polynomials in terms of a simple geometric structure of the sampling pattern, and therefore we obtain the deterministic necessary and sufficient condition on the sampling pattern for finite completability of the sampled tensor. Moreover, assuming that the entries of the tensor are sampled independently with probability $p$ and using the mentioned deterministic analysis, we propose a combinatorial method to derive a lower bound on the sampling probability $p$, or equivalently, the number of sampled entries that guarantees finite completability with high probability. We also show that the existing result for the matrix completion problem can be used to obtain a loose lower bound on the sampling probability $p$. In addition, we obtain deterministic and probabilistic conditions for unique completability. It is seen that the number of samples required for finite or unique completability obtained by the proposed analysis on the CP manifold is orders-of-magnitude lower than that is obtained by the existing analysis on the Grassmannian manifold.

Rating platforms enable large-scale collection of user opinion about items (products, other users, etc.). However, many untrustworthy users give fraudulent ratings for excessive monetary gains. In the paper, we present FairJudge, a system to identify such fraudulent users. We propose three metrics: (i) the fairness of a user that quantifies how trustworthy the user is in rating the products, (ii) the reliability of a rating that measures how reliable the rating is, and (iii) the goodness of a product that measures the quality of the product. Intuitively, a user is fair if it provides reliable ratings that are close to the goodness of the product. We formulate a mutually recursive definition of these metrics, and further address cold start problems and incorporate behavioral properties of users and products in the formulation. We propose an iterative algorithm, FairJudge, to predict the values of the three metrics. We prove that FairJudge is guaranteed to converge in a bounded number of iterations, with linear time complexity. By conducting five different experiments on five rating platforms, we show that FairJudge significantly outperforms nine existing algorithms in predicting fair and unfair users. We reported the 100 most unfair users in the Flipkart network to their review fraud investigators, and 80 users were correctly identified (80% accuracy). The FairJudge algorithm is already being deployed at Flipkart.

This paper proposes to improve visual question answering (VQA) with structured representations of both scene contents and questions. A key challenge in VQA is to require joint reasoning over the visual and text domains. The predominant CNN/LSTM-based approach to VQA is limited by monolithic vector representations that largely ignore structure in the scene and in the form of the question. CNN feature vectors cannot effectively capture situations as simple as multiple object instances, and LSTMs process questions as series of words, which does not reflect the true complexity of language structure. We instead propose to build graphs over the scene objects and over the question words, and we describe a deep neural network that exploits the structure in these representations. This shows significant benefit over the sequential processing of LSTMs. The overall efficacy of our approach is demonstrated by significant improvements over the state-of-the-art, from 71.2% to 74.4% in accuracy on the "abstract scenes" multiple-choice benchmark, and from 34.7% to 39.1% in accuracy over pairs of "balanced" scenes, i.e. images with fine-grained differences and opposite yes/no answers to a same question.

The ability to predict how an environment changes based on forces applied to it is fundamental for a robot to achieve specific goals. For instance, in order to arrange objects on a table into a desired configuration, a robot has to be able to reason about where and how to push individual objects, which requires some understanding of physical quantities such as object boundaries, mass, surface friction, and their relationship to forces. A standard approach in robot control is to use a physical model of the environment and perform optimal control to find a policy that leads to the goal state. For instance, extensive work utilizing the MuJoCo physics engine [1] has shown how strong physics models can enable solutions to control problems in complex and contact-rich environments [2]. A shortcoming of such models is, however, that they rely on very accurate estimates of the state of the system [3].

Nonparametric models are versatile, albeit computationally expensive, tool for modeling mixture models. In this paper, we introduce spectral methods for the two most popular nonparametric models: the Indian Buffet Process (IBP) and the Hierarchical Dirichlet Process (HDP). We show that using spectral methods for the inference of nonparametric models are computationally and statistically efficient. In particular, we derive the lower-order moments of the IBP and the HDP, propose spectral algorithms for both models, and provide reconstruction guarantees for the algorithms. For the HDP, we further show that applying hierarchical models on dataset with hierarchical structure, which can be solved with the generalized spectral HDP, produces better solutions to that of flat models regarding likelihood performance.

Artificial neural networks (ANNs) have gained a well-deserved popularity among machine learning tools upon their recent successful applications in image- and sound processing and classification problems. ANNs have also been applied for predicting the family or function of a protein, knowing its residue sequence. Here we present two new ANNs with multi-label classification ability, showing impressive accuracy when classifying protein sequences into 698 UniProt families (AUC=99.99%) and 983 Gene Ontology classes (AUC=99.45%).

Empirical scoring functions based on either molecular force fields or cheminformatics descriptors are widely used, in conjunction with molecular docking, during the early stages of drug discovery to predict potency and binding affinity of a drug-like molecule to a given target. These models require expert-level knowledge of physical chemistry and biology to be encoded as hand-tuned parameters or features rather than allowing the underlying model to select features in a data-driven procedure. Here, we develop a general 3-dimensional spatial convolution operation for learning atomic-level chemical interactions directly from atomic coordinates and demonstrate its application to structure-based bioactivity prediction. The atomic convolutional neural network is trained to predict the experimentally determined binding affinity of a protein-ligand complex by direct calculation of the energy associated with the complex, protein, and ligand given the crystal structure of the binding pose. Non-covalent interactions present in the complex that are absent in the protein-ligand substructures are identified and the model learns the interaction strength associated with these features. We test our model by predicting the binding free energy of a subset of protein-ligand complexes found in the PDBBind dataset and compare with state-of-the-art cheminformatics and machine learning-based approaches. We find that all methods achieve experimental accuracy (less than 1 kcal/mol mean absolute error) and that atomic convolutional networks either outperform or perform competitively with the cheminformatics based methods. Unlike all previous protein-ligand prediction systems, atomic convolutional networks are end-to-end and fully-differentiable. They represent a new data-driven, physics-based deep learning model paradigm that offers a strong foundation for future improvements in structure-based bioactivity prediction.

We consider the problems of robust PAC learning from distributed and streaming data, which may contain malicious errors and outliers, and analyze their fundamental complexity questions. In particular, we establish lower bounds on the communication complexity for distributed robust learning performed on multiple machines, and on the space complexity for robust learning from streaming data on a single machine. These results demonstrate that gaining robustness of learning algorithms is usually at the expense of increased complexities. As far as we know, this work gives the first complexity results for distributed and online robust PAC learning.