One of the central goals of Recurrent Neural Networks (RNNs) is to learn long-term dependencies in sequential data. Nevertheless, the most popular training method, Truncated Backpropagation through Time (TBPTT), categorically forbids learning dependencies beyond the truncation horizon. In contrast, the online training algorithm Real Time Recurrent Learning (RTRL) provides untruncated gradients, with the disadvantage of impractically large computational costs. Recently published approaches reduce these costs by providing noisy approximations of RTRL. We present a new approximation algorithm of RTRL, Optimal Kronecker-Sum Approximation (OK). We prove that OK is optimal for a class of approximations of RTRL, which includes all approaches published so far. Additionally, we show that OK has empirically negligible noise: Unlike previous algorithms it matches TBPTT in a real world task (character-level Penn TreeBank) and can exploit online parameter updates to outperform TBPTT in a synthetic string memorization task.

Complex networks are used as an abstraction for systems modeling in physics, biology, sociology, and other areas. We propose an algorithm based on a fast personalized node ranking and recent advancements in deep learning for learning supervised network embeddings as well as to classify network nodes directly. Learning from homogeneous, as well as heterogeneous networks, our algorithm outperforms strong baselines on nine node-classification benchmarks from the domains of molecular biology, finance, social media and language processing---one of the largest node classification collections to date. The results are comparable or better than current state-of-the-art in terms of speed as well as predictive accuracy. Embeddings, obtained by the proposed algorithm, are also a viable option for network visualization.

The posteriors over neural network weights are high dimensional and multimodal. Each mode typically characterizes a meaningfully different representation of the data. We develop Cyclical Stochastic Gradient MCMC (SG-MCMC) to automatically explore such distributions. In particular, we propose a cyclical stepsize schedule, where larger steps discover new modes, and smaller steps characterize each mode. We prove that our proposed learning rate schedule provides faster convergence to samples from a stationary distribution than SG-MCMC with standard decaying schedules. Moreover, we provide extensive experimental results to demonstrate the effectiveness of cyclical SG-MCMC in learning complex multimodal distributions, especially for fully Bayesian inference with modern deep neural networks.

In large scale systems, approximate nearest neighbour search is a crucial algorithm to enable efficient data retrievals. Recently, deep learning-based hashing algorithms have been proposed as a promising paradigm to enable data dependent schemes. Often their efficacy is only demonstrated on data sets with fixed, limited numbers of classes. In practical scenarios, those labels are not always available or one requires a method that can handle a higher input variability, as well as a higher granularity. To fulfil those requirements, we look at more flexible similarity measures. In this work, we present a novel, flexible, end-to-end trainable network for large-scale data hashing. Our method works by transforming the data distribution to behave as a uniform distribution on a product of spheres. The transformed data is subsequently hashed to a binary form in a way that maximises entropy of the output, (i.e. to fully utilise the available bit-rate capacity) while maintaining the correctness (i.e. close items hash to the same key in the map). We show that the method outperforms baseline approaches such as locality-sensitive hashing and product quantisation in the limited capacity regime.

Capsule network (CapsNet) was introduced as an enhancement over convolutional neural networks, supplementing the latter's invariance properties with equivariance through pose estimation. CapsNet achieved a very decent performance with a shallow architecture and a significant reduction in parameters count. However, the width of the first layer in CapsNet is still contributing to a significant number of its parameters and the shallowness may be limiting the representational power of the capsules. To address these limitations, we introduce Path Capsule Network (PathCapsNet), a deep parallel multi-path version of CapsNet. We show that a judicious coordination of depth, max-pooling, regularization by DropCircuit and a new fan-in routing by agreement technique can achieve better or comparable results to CapsNet, while further reducing the parameter count significantly.

Adversarial examples are perturbed inputs that are designed (from a deep learning network's (DLN) parameter gradients) to mislead the DLN during test time. Intuitively, constraining the dimensionality of inputs or parameters of a network reduces the 'space' in which adversarial examples exist. Guided by this intuition, we demonstrate that discretization greatly improves the robustness of DLNs against adversarial attacks. Specifically, discretizing the input space (or allowed pixel levels from 256 values or 8-bit to 4 values or 2-bit) extensively improves the adversarial robustness of DLNs for a substantial range of perturbations for minimal loss in test accuracy. Furthermore, we find that Binary Neural Networks (BNNs) and related variants are intrinsically more robust than their full precision counterparts in adversarial scenarios. Combining input discretization with BNNs furthers the robustness even waiving the need for adversarial training for certain magnitude of perturbation values. We evaluate the effect of discretization on MNIST, CIFAR10, CIFAR100 and Imagenet datasets. Across all datasets, we observe maximal adversarial resistance with 2-bit input discretization that incurs an adversarial accuracy loss of just ~1-2% as compared to clean test accuracy.

Crimes emerge out of complex interactions of behaviors and situations; thus there are complex linkages between crime incidents. Solving the puzzle of crime linkage is a highly challenging task because we often only have limited information from indirect observations such as records, text descriptions, and associated time and locations. We propose a new modeling and learning framework for detecting linkage between crime events using \textit{spatio-temporal-textual} data, which are highly prevalent in the form of police reports. We capture the notion of \textit{modus operandi} (M.O.), by introducing a multivariate marked point process and handling the complex text jointly with the time and location. The model is able to discover the latent space that links the crime series. The model fitting is achieved by a computationally efficient Expectation-Maximization (EM) algorithm. In addition, we explicitly reduce the bias in the text documents in our algorithm. Our numerical results using real data from the Atlanta Police show that our method has competitive performance relative to the state-of-the-art. Our results, including variable selection, are highly interpretable and may bring insights into M.O. extraction.

Predicting properties of nodes in a graph is an important problem with applications in a variety of domains. Graph-based Semi-Supervised Learning (SSL) methods aim to address this problem by labeling a small subset of the nodes as seeds and then utilizing the graph structure to predict label scores for the rest of the nodes in the graph. Recently, Graph Convolutional Networks (GCNs) have achieved impressive performance on the graph-based SSL task. In addition to label scores, it is also desirable to have confidence scores associated with them. Unfortunately, confidence estimation in the context of GCN has not been previously explored. We fill this important gap in this paper and propose ConfGCN, which estimates labels scores along with their confidences jointly in GCN-based setting. ConfGCN uses these estimated confidences to determine the influence of one node on another during neighborhood aggregation, thereby acquiring anisotropic capabilities. Through extensive analysis and experiments on standard benchmarks, we find that ConfGCN is able to outperform state-of-the-art baselines. We have made ConfGCN's source code available to encourage reproducible research.

A deep learning model is proposed for predicting block-level parking occupancy in real time. The model leverages Graph-Convolutional Neural Networks (GCNN) to extract the spatial relations of traffic flow in large-scale networks, and utilizes Recurrent Neural Networks (RNN) with Long-Short Term Memory (LSTM) to capture the temporal features. In addition, the model is capable of taking multiple heterogeneously structured traffic data sources as input, such as parking meter transactions, traffic speed, and weather conditions. The model performance is evaluated through a case study in Pittsburgh downtown area. The proposed model outperforms other baseline methods including multi-layer LSTM and Lasso with an average testing MAPE of 12.0\% when predicting block-level parking occupancies 30 minutes in advance. The case study also shows that, in generally, the prediction model works better for business areas than for recreational locations. We found that incorporating traffic speed and weather information can significantly improve the prediction performance. Weather data is particularly useful for improving predicting accuracy in recreational areas.

It's moving fast with new research coming out each and every day allowing us to do truly amazing things that we could't do before with computers and AI. The figure below shows results from Mask-RCNN, one of the most flexible and powerful deep networks ever designed for computer vision. It's capable of generating the phenomenal segmentations you see below. Convolutional Neural Networks (CNNs) are the driving force behind every advancement in Computer Vision research we see today. Most people know how they work in a general sense.