In this work, we contribute a new multi-layer neural network architecture named ONCF to perform collaborative filtering. The idea is to use an outer product to explicitly model the pairwise correlations between the dimensions of the embedding space. In contrast to existing neural recommender models that combine user embedding and item embedding via a simple concatenation or element-wise product, our proposal of using outer product above the embedding layer results in a two-dimensional interaction map that is more expressive and semantically plausible. Above the interaction map obtained by outer product, we propose to employ a convolutional neural network to learn high-order correlations among embedding dimensions. Extensive experiments on two public implicit feedback data demonstrate the effectiveness of our proposed ONCF framework, in particular, the positive effect of using outer product to model the correlations between embedding dimensions in the low level of multi-layer neural recommender model. The experiment codes are available at: https://github.com/duxy-me/ConvNCF

Item recommendation is a personalized ranking task. To this end, many recommender systems optimize models with pairwise ranking objectives, such as the Bayesian Personalized Ranking (BPR). Using matrix Factorization (MF) --- the most widely used model in recommendation --- as a demonstration, we show that optimizing it with BPR leads to a recommender model that is not robust. In particular, we find that the resultant model is highly vulnerable to adversarial perturbations on its model parameters, which implies the possibly large error in generalization. To enhance the robustness of a recommender model and thus improve its generalization performance, we propose a new optimization framework, namely Adversarial Personalized Ranking (APR). In short, our APR enhances the pairwise ranking method BPR by performing adversarial training. It can be interpreted as playing a minimax game, where the minimization of the BPR objective function meanwhile defends an adversary, which adds adversarial perturbations on model parameters to maximize the BPR objective function. To illustrate how it works, we implement APR on MF by adding adversarial perturbations on the embedding vectors of users and items. Extensive experiments on three public real-world datasets demonstrate the effectiveness of APR --- by optimizing MF with APR, it outperforms BPR with a relative improvement of 11.2% on average and achieves state-of-the-art performance for item recommendation. Our implementation is available at: https://github.com/hexiangnan/adversarial_personalized_ranking.

Neuronal circuits formed in the brain are complex with intricate connection patterns. Such a complexity is also observed in the retina as a relatively simple neuronal circuit. A retinal ganglion cell receives excitatory inputs from neurons in previous layers as driving forces to fire spikes. Analytical methods are required that can decipher these components in a systematic manner. Recently a method termed spike-triggered non-negative matrix factorization (STNMF) has been proposed for this purpose. In this study, we extend the scope of the STNMF method. By using the retinal ganglion cell as a model system, we show that STNMF can detect various biophysical properties of upstream bipolar cells, including spatial receptive fields, temporal filters, and transfer nonlinearity. In addition, we recover synaptic connection strengths from the weight matrix of STNMF. Furthermore, we show that STNMF can separate spikes of a ganglion cell into a few subsets of spikes where each subset is contributed by one presynaptic bipolar cell. Taken together, these results corroborate that STNMF is a useful method for deciphering the structure of neuronal circuits.

Welcome everyone to an updated deep learning with Python and Tensorflow tutorial mini-series. Since doing the first deep learning with TensorFlow course a little over 2 years ago, much has changed. It's nowhere near as complicated to get started, nor do you need to know as much to be successful with deep learning. If you're interested in more of the details with how TensorFlow works, you can still check out the previous tutorials, as they go over the more raw TensorFlow. This is more of a deep learning quick start!

Paul Zikopoulos, vice president of Big Data & Cognitive Systems, IBM speaks during the annual Tallahassee Chamber Conference at the Omni Amelia Island Plantation on Saturday, Aug. 11, 2018. AMELIA ISLAND – Tech years are a lot like dog years -- they go very quickly, said one of the nation's top Big Data experts during the Tallahassee Chamber of Commerce Conference on Saturday. Companies can transform their delivery services and business models by the "one percent rule" over time and reap catalytic results, said Paul Zikopoulos, vice president of Big Data and Cognitive Systems for IBM. Zikopoulos is the author of 19 books and a leading authority on Big Data. According to Leading Authorities Inc., he was named one of the "50 Big Data Twitter Influencers" by SAP, Zikopoulos has served as a consultant for 60 Minutes and multiple universities and has been named an expert on big data by publications such as Big Data Republic, Technopedia, and Analytics Week.

Data from multiple sources, such as weather reports, satellite images, disaster history, can be used to train a deep learning system. With insights drawn from thorough analysis, experts can predict the imminent occurrence of disasters, helping people to follow a proactive approach and minimize the impact of catastrophes. No one can pause or halt the occurrence of natural or human-made disasters. We can only take steps to reduce their impact. By integrating deep learning applications such as drones, experts can get real-time data on areas that are about to get hit by a disaster.

Here's how it's related to artificial intelligence, how it works and why it matters. Deep learning is a subset of machine learning, which itself falls within the field of artificial intelligence. Artificial intelligence is the study of how to build machines capable of carrying out tasks that would typically require human intelligence. That rather loose definition means that AI encompasses many fields of research, from genetic algorithms to expert systems, and provides scope for arguments over what constitutes AI. Within the field of AI research, machine learning has enjoyed remarkable success in recent years -- allowing computers to surpass or come close to matching human performance in areas ranging from facial recognition to speech and language recognition. Machine learning is the process of teaching a computer to carry out a task, rather than programming it how to carry that task out step by step. At the end of training, a machine-learning system will be able to make accurate predictions when given data.

Note from Jeremy: I'll be teaching Deep Learning for Coders at the University of San Francisco starting in October; if you've got at least a year of coding experience, you can apply here. This is a new speed record for training Imagenet to this accuracy on publicly available infrastructure, and is 40% faster than Google's DAWNBench record on their proprietary TPU Pod cluster. Our approach uses the same number of processing units as Google's benchmark (128) and costs around $40 to run. The main training methods we used (details below) are: fast.ai's We used the classic ResNet-50 architecture, and SGD with momentum.

Researchers often try to capture as much information as they can, either by using existing architectures, creating new ones, going deeper, or employing different training methods. This paper compares different ideas and methods that are used heavily in Machine Learning to determine what works best. These methods are prevalent in various domains of Machine Learning, such as Computer Vision and Natural Language Processing (NLP). Throughout our work, we have tried to bring generalization into context, because that's what matters in the end. Any model should be robust and able to work outside your research environment. When a model lacks generalization, very often we try to train the model on datasets it has never encountered … and that's when things start to get much more complex.

Thousands of leading scientists have urged their colleagues against helping governments create killer robots, making the movie The Terminator a reality. In other words, killer robots, that could eventually develop a mind of their own and take over the world. Scientists have vowed they will not support robots "that can identify and attack people without human oversight." Two leading experts backing the commitment, Demis Hassabis at Google DeepMind and Elon Musk at SpaceX, are among the more than 2,400 signatories whom have pledged to deter military firms and nations from building lethal autonomous weapon systems, referred to as Laws. The move is the latest from concerned scientists and organizations about giving a machine the power to choose someone's fate of life or death.