The goal of a recommendation system is to predict the interest of a user in a given item by exploiting the existing set of ratings as well as certain user/item features. A standard approach to modeling this problem is Inductive Matrix Completion where the predicted rating is modeled as an inner product of the user and the item features projected onto a latent space. In order to learn the parameters effectively from a small number of observed ratings, the latent space is constrained to be low-dimensional which implies that the parameter matrix is constrained to be low-rank. However, such bilinear modeling of the ratings can be limiting in practice and non-linear prediction functions can lead to significant improvements. A natural approach to introducing non-linearity in the prediction function is to apply a non-linear activation function on top of the projected user/item features. Imposition of non-linearities further complicates an already challenging problem that has two sources of non-convexity: a) low-rank structure of the parameter matrix, and b) non-linear activation function. We show that one can still solve the non-linear Inductive Matrix Completion problem using gradient descent type methods as long as the solution is initialized well. That is, close to the optima, the optimization function is strongly convex and hence admits standard optimization techniques, at least for certain activation functions, such as Sigmoid and tanh. We also highlight the importance of the activation function and show how ReLU can behave significantly differently than say a sigmoid function. Finally, we apply our proposed technique to recommendation systems and semi-supervised clustering, and show that our method can lead to much better performance than standard linear Inductive Matrix Completion methods.

The focus in machine learning has branched beyond training classifiers on a single task to investigating how previously acquired knowledge in a source domain can be leveraged to facilitate learning in a related target domain, known as inductive transfer learning. Three active lines of research have independently explored transfer learning using neural networks. In weight transfer, a model trained on the source domain is used as an initialization point for a network to be trained on the target domain. In deep metric learning, the source domain is used to construct an embedding that captures class structure in both the source and target domains. In few-shot learning, the focus is on generalizing well in the target domain based on a limited number of labeled examples. We compare state-of-the-art methods from these three paradigms and also explore hybrid adapted-embedding methods that use limited target-domain data to fine tune embeddings constructed from source-domain data. We conduct a systematic comparison of methods in a variety of domains, varying the number of labeled instances available in the target domain ($k$), as well as the number of target-domain classes. We reach three principal conclusions: (1) Deep embeddings are far superior, compared to weight transfer, as a starting point for inter-domain transfer or model re-use (2) Our hybrid methods robustly outperform every few-shot learning and every deep metric learning method previously proposed, with a mean error reduction of 30% over state-of-the-art. (3) Among loss functions for discovering embeddings, the histogram loss (Ustinova & Lempitsky, 2016) is most robust. We hope our results will motivate a unification of research in weight transfer, deep metric learning, and few-shot learning.

Graph Convolutional Neural Networks (GCNNs) are the most recent exciting advancement in deep learning field and their applications are quickly spreading in multi-cross-domains including bioinformatics, chemoinformatics, social networks, natural language processing and computer vision. In this paper, we expose and tackle some of the basic weaknesses of a GCNN model with a capsule idea presented in \cite{hinton2011transforming} and propose our Graph Capsule Network (GCAPS-CNN) model. In addition, we design our GCAPS-CNN model to solve especially graph classification problem which current GCNN models find challenging. Through extensive experiments, we show that our proposed Graph Capsule Network can significantly outperforms both the existing state-of-art deep learning methods and graph kernels on graph classification benchmark datasets.

A variety of machine learning models have been proposed to assess the performance of players in professional sports. However, they have only a limited ability to model how player performance depends on the game context. This paper proposes a new approach to capturing game context: we apply Deep Reinforcement Learning (DRL) to learn an action-value Q function from 3M play-by-play events in the National Hockey League (NHL). The neural network representation integrates both continuous context signals and game history, using a possession-based LSTM. The learned Q-function is used to value players' actions under different game contexts. To assess a player's overall performance, we introduce a novel Game Impact Metric (GIM) that aggregates the values of the player's actions. Empirical Evaluation shows GIM is consistent throughout a play season, and correlates highly with standard success measures and future salary.

We present a novel deep learning architecture to address the cloze-style question answering task. Existing approaches employ reading mechanisms that do not fully exploit the interdependency between the document and the query. In this paper, we propose a novel \emph{dependent gated reading} bidirectional GRU network (DGR) to efficiently model the relationship between the document and the query during encoding and decision making. Our evaluation shows that DGR obtains highly competitive performance on well-known machine comprehension benchmarks such as the Children's Book Test (CBT-NE and CBT-CN) and Who DiD What (WDW, Strict and Relaxed). Finally, we extensively analyze and validate our model by ablation and attention studies.

In this work we design and train deep neural networks to predict topological invariants for one-dimensional four-band insulators in AIII class whose topological invariant is the winding number, and two-dimensional two-band insulators in A class whose topological invariant is the Chern number. Given Hamiltonians in the momentum space as the input, neural networks can predict topological invariants for both classes with accuracy close to or higher than 90%, even for Hamiltonians whose invariants are beyond the training data set. Despite the complexity of the neural network, we find that the output of certain intermediate hidden layers resembles either the winding angle for models in AIII class or the solid angle (Berry curvature) for models in A class, indicating that neural networks essentially capture the mathematical formula of topological invariants. Our work demonstrates the ability of neural networks to predict topological invariants for complicated models with local Hamiltonians as the only input, and offers an example that even a deep neural network is understandable.

Artificial Intelligence is the future. For a common man, artificial intelligence is a science fiction and is already a part and parcel of our daily lives. Machine learning and deep learning are the two buzzing terms associated with artificial intelligence. Deep learning, machine learning and artificial intelligence are a set of Russian dolls nested with each other beginning with the smallest and working out. This article will help in understanding artificial intelligence, machine learning and deep learning and the difference among them.

AI has already proven to be pretty adroit at spotting early signs of disease in medical data. The latest project, from researchers at Georgia Tech, shows the potential of AI to spot early signs of heart failure. The research, which appears in the Journal of the American Medical Informatics Association (JAMIA), utilizes deep learning to examine data for temporality to provide earlier detections of incidents that can often lead to heart failure. It's an approach that is relatively unique in the application of artificial intelligence in healthcare, yet the researchers believe it could allow for much earlier detection of heart issues by identifying incidents that usually result in heart failure up to 18 months into the future. They can achieve this by looking for temporal relations among events that appear in our electronic medical records. It's an approach that is often used in natural language processing, but much less so in deep learning.

Qualcomm and Baidu are deepening their partnership on Artificial Intelligence, announcing that they will essentially combine their Qualcomm Artificial Intelligence and Baidu PaddlePaddle platforms. The latter was first launched as a deep learning platform for internal use at Baidu in 2013, with the company subsequently making it open source in August of 2016. And with Baidu's announcement in February of this year that it would support the Qualcomm AI Engine, bringing Baidu PaddlePaddle into the mix appears to be a logical next step. The companies will combine their technologies through the Open Neural Network Exchange ("ONNX"), an open source platform aimed at allowing developers to easily choose between a number of different tools and models as they build AI technologies. Other partners of the ONNX include Microsoft, Facebook, NVIDIA, and Amazon Web Services.

Edureka's Masters Program is a thoughtful compilation of Instructor -Led and Self Paced Courses, allowing the learners to be guided by industry experts, as well as learn skills at their own pace. In the Data Science Masters Program, Data Science Certification Course using R, Python Certification Training for Data Science, Apache Spark and Scala Certification Training, AI & Deep Learning with TensorFlow, Tableau Training & Certification are Instructor - led Online Courses.