A temporal point process is a mathematical model for a time series of discrete events, which covers various applications. Recently, recurrent neural network (RNN) based models have been developed for point processes and have been found effective. RNN based models usually assume a specific functional form for the time course of the intensity function of a point process (e.g., exponentially decreasing or increasing with the time since the most recent event). However, such an assumption can restrict the expressive power of the model. We herein propose a novel RNN based model in which the time course of the intensity function is represented in a general manner. In our approach, we first model the integral of the intensity function using a feedforward neural network and then obtain the intensity function as its derivative. This approach enables us to both obtain a flexible model of the intensity function and exactly evaluate the log-likelihood function, which contains the integral of the intensity function, without any numerical approximations. Our model achieves competitive or superior performances compared to the previous state-of-the-art methods for both synthetic and real datasets.

Predicting interactions between structured entities lies at the core of numerous tasks such as drug regimen and new material design. In recent years, graph neural networks have become attractive. They represent structured entities as graphs and then extract features from each individual graph using graph convolution operations. However, these methods have some limitations: i) their networks only extract features from a fix-sized subgraph structure (i.e., a fix-sized receptive field) of each node, and ignore features in substructures of different sizes, and ii) features are extracted by considering each entity independently, which may not effectively reflect the interaction between two entities. To resolve these problems, we present MR-GNN, an end-to-end graph neural network with the following features: i) it uses a multi-resolution based architecture to extract node features from different neighborhoods of each node, and, ii) it uses dual graph-state long short-term memory networks (L-STMs) to summarize local features of each graph and extracts the interaction features between pairwise graphs. Experiments conducted on real-world datasets show that MR-GNN improves the prediction of state-of-the-art methods.

In recent years, deep learning has achieved remarkable achievements in many fields, including computer vision, natural language processing, speech recognition and others. Adequate training data is the key to ensure the effectiveness of the deep models. However, obtaining valid data requires a lot of time and labor resources. Data augmentation (DA) is an effective alternative approach, which can generate new labeled data based on existing data using label-preserving transformations. Although we can benefit a lot from DA, designing appropriate DA policies requires a lot of expert experience and time consumption, and the evaluation of searching the optimal policies is costly. So we raise a new question in this paper: how to achieve automated data augmentation at as low cost as possible? We propose a method named BO-Aug for automating the process by finding the optimal DA policies using the Bayesian optimization approach. Our method can find the optimal policies at a relatively low search cost, and the searched policies based on a specific dataset are transferable across different neural network architectures or even different datasets. We validate the BO-Aug on three widely used image classification datasets, including CIFAR-10, CIFAR-100 and SVHN. Experimental results show that the proposed method can achieve state-of-the-art or near advanced classification accuracy. Code to reproduce our experiments is available at https://github.com/zhangxiaozao/BO-Aug.

Geoffrey Hinton is one of the creators of Deep Learning, a 2019 winner of the Turing Award, and an engineering fellow at Google. Last week, at the company's I/O developer conference, we discussed his early fascination with the brain, and the possibility that computers could be modeled after its neural structure--an idea long dismissed by other scholars as foolhardy. We also discussed consciousness, his future plans, and whether computers should be taught to dream. The conversation has been lightly edited for length and clarity. Nicholas Thompson: Let's start when you write some of your early, very influential papers. Everybody says, "This is a smart idea, but we're not actually going to be able to design computers this way." Explain why you persisted and why you were so confident that you had found something important. Geoffrey Hinton: It seemed to me there's no other way the brain could work. It has to work by learning the strength of connections. And if you want to make a device do something intelligent, you've got two options: You can program it, or it can learn. And people certainly weren't programmed, so we had to learn. This had to be the right way to go. NT: Explain what neural networks are. GH: You have relatively simple processing elements that are very loosely models of neurons. They have connections coming in, each connection has a weight on it, and that weight can be changed through learning. And what a neuron does is take the activities on the connections times the weights, adds them all up, and then decides whether to send an output.

With the rise of libraries such as Tensorflow 2.0, PyTorch and Fastai, implementing deep learning has become accessible to so many more people and it helps to understand the fundamentals behind deep neural networks. Hopefully this article will be of help people to people on the path of learning about deep neural networks. Back when I first learnt about neural nets and implemented my first, they were always represented as individual artificial neurons, essentially nodes with individually weighted inputs, a summed output and an activation function. When first returning into learning about deep neural networks, the concept of how this equated to matrix multiplication didn't appear obvious. Also, linked to this is why Graphics Processing Units (GPUs) and their spin-offs have helped advance deep learning results so much.

Let's dig deeper into the details of the algorithm. The authors used a Weisfeiler-Lehman Network, a type of graph-convolutional neural network, as depicted in the figure below. First, for each atom-atom pair (including pairs of atoms that are not bound or are located in different reactants) the neural network predicts the likelihood of the bond order to change. The starting point is from a graph representation of reactants (A), where atoms are featurized by atomic number, formal charge, degree of connectivity, valence and aromaticity, and bonds are featurized by bond order and ring status. These atom-level features are iteratively updated (B) by incorporating information from neighbor atoms.

You can follow any comments to this entry through the RSS 2.0 feed. You can skip to the end and leave a comment. Pinging is currently not allowed. Your email address will not be published. This site uses Akismet to reduce spam.

One of the most interesting demos at this week's Google I/O keynote featured a new version of Google's voice assistant that's due out later this year. A Google employee asked the Google Assistant to bring up her photos and then show her photos with animals. She tapped one and said, "Send it to Justin." The photo was dropped into the messaging app. From there, things got more impressive.

The model is significantly faster to train and to make predictions, yet still requires a set of candidate regions to be proposed along with each input image. Python and C (Caffe) source code for Fast R-CNN as described in the paper was made available in a GitHub repository. The model architecture was further improved for both speed of training and detection by Shaoqing Ren, et al. at Microsoft Research in the 2016 paper titled "Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks." The architecture was the basis for the first-place results achieved on both the ILSVRC-2015 and MS COCO-2015 object recognition and detection competition tasks. The architecture was designed to both propose and refine region proposals as part of the training process, referred to as a Region Proposal Network, or RPN. These regions are then used in concert with a Fast R-CNN model in a single model design. These improvements both reduce the number of region proposals and accelerate the test-time operation of the model to near real-time with then state-of-the-art performance.

For those who like their dessert first: here's the finished model, and here's the colab for this example. A rather empty user-interface should show up on your screen. In the sidebar, click the Library-dropdown, and select TensorFlow. Now the code for our model will use TensorFlow instead of PyTorch. Next, click on the Theme-dropdown and select "orange".