Deep Neural Network (DNN) has been largely demonstrated to be effective for closed-world classification problems where the total number of classes are known in advance. However, when the total number of classes that may occur during test time is unknown, DNNs notorious fail, i.e., DNN will make incorrect label prediction on instances from novel or unseen classes. This severely limits its utility in many real-world web applications, particularly when data occurs as a continuous stream. In this paper, we focus on addressing this key challenge by developing a two-channel DNN based co-representation learning framework that not only predicts instances from known classes, but also detects and adapts to the occurrence of novel class instances over time. Concretely, we propose a metric learning method using pairwise-constraint loss (PCL) function to learn a feature representation where intra-class compactness and inter-class separation is achieved. Moreover, we apply the temperature scaling scheme on the softmax function to replace traditional softmax output and design an open-world classifier. Our extensive empirical evaluation on benchmark datasets demonstrates the effectiveness of our framework compared to other competing techniques.

We detail a new framework for privacy preserving deep learning and discuss its assets. The framework puts a premium on ownership and secure processing of data and introduces a valuable representation based on chains of commands and tensors. This abstraction allows one to implement complex privacy preserving constructs such as Federated Learning, Secure Multiparty Computation, and Differential Privacy while still exposing a familiar deep learning API to the end-user. We report early results on the Boston Housing and Pima Indian Diabetes datasets. While the privacy features apart from Differential Privacy do not impact the prediction accuracy, the current implementation of the framework introduces a significant overhead in performance, which will be addressed at a later stage of the development. We believe this work is an important milestone introducing the first reliable, general framework for privacy preserving deep learning.

We introduce GSimCNN (Graph Similarity Computation via Convolutional Neural Networks) for predicting the similarity score between two graphs. As the core operation of graph similarity search, pairwise graph similarity computation is a challenging problem due to the NPhard nature of computing many graph distance/similarity metrics. We demonstrate our model using the Graph Edit Distance (GED) [2] as the example metric. It is defined as the number of edit operations in the optimal alignments that transform one graph into the other, where an edit operation can be an insertion or a deletion of a node/edge, or relabelling of a node. It is NPhard [3] and costly to compute in practice [4]. The key idea is to turn the pairwise graph distance computation problem into a learning problem.

Despite showing state-of-the-art performance, deep learning for speech recognition remains challenging to deploy in on-device edge scenarios such as mobile and other consumer devices. Recently, there have been greater efforts in the design of small, low-footprint deep neural networks (DNNs) that are more appropriate for edge devices, with much of the focus on design principles for hand-crafting efficient network architectures. In this study, we explore a human-machine collaborative design strategy for building low-footprint DNN architectures for speech recognition through a marriage of human-driven principled network design prototyping and machine-driven design exploration. The efficacy of this design strategy is demonstrated through the design of a family of highly-efficient DNNs (nicknamed EdgeSpeechNets) for limited-vocabulary speech recognition. Experimental results using the Google Speech Commands dataset for limited-vocabulary speech recognition showed that EdgeSpeechNets have higher accuracies than state-of-the-art DNNs (with the best EdgeSpeechNet achieving ~97% accuracy), while achieving significantly smaller network sizes (as much as 7.8x smaller) and lower computational cost (as much as 36x fewer multiply-add operations, 10x lower prediction latency, and 16x smaller memory footprint on a Motorola Moto E phone), making them very well-suited for on-device edge voice interface applications.

Neural message passing algorithms for semi-supervised classification on graphs have recently achieved great success. However, these methods only consider nodes that are a few propagation steps away and the size of this utilized neighborhood cannot be easily extended. In this paper, we use the relationship between graph convolutional networks (GCN) and PageRank to derive an improved propagation scheme based on personalized PageRank. We utilize this propagation procedure to construct personalized propagation of neural predictions (PPNP) and its approximation, APPNP. Our model's training time is on par or faster and its number of parameters on par or lower than previous models. It leverages a large, adjustable neighborhood for classification and can be combined with any neural network. We show that this model outperforms several recently proposed methods for semi-supervised classification on multiple graphs in the most thorough study done so far for GCN-like models. Graphs are ubiquitous in the real world and its description through scientific models. They are used to study the spread of information, to optimize delivery, to recommend new books, to suggest friends, or to find a party's potential voters.

Abstract-- The energy output of photovoltaic (PV) power plants depends on the environment and thus fluctuates over time. As a result, PV power can cause instability in the power grid, in particular when increasingly used. Limiting the rate of change of the power output is a common way to mitigate these fluctuations, often with the help of large batteries. A reactive controller that uses these batteries to compensate ramps works in practice, but causes stress on the battery due to a high energy throughput. In this paper, we present a deep learning approach that uses images of the sky to compensate power fluctuations predictively and reduces battery stress. In particular, we show that the optimal control policy can be computed using information that is only available in hindsight. Based on this, we use imitation learning to train a neural network that approximates this hindsight-optimal policy, but uses only currently available sky images and sensor data. We evaluate our method on a large dataset of measurements and images from a real power plant and show that the trained policy reduces stress on the battery. Photovoltaic (PV) power generation has grown at a rate of roughly 30% per year in recent years and reached a global capacity of over 400 GW at the end of 2017 [1].

Networks are fundamental building blocks for representing data, and computations. Remarkable progress in learning in structurally defined (shallow or deep) networks has recently been achieved. Here we introduce evolutionary exploratory search and learning method of topologically flexible networks under the constraint of producing elementary computational steady-state input-output operations. Our results include; (1) the identification of networks, over four orders of magnitude, implementing computation of steady-state input-output functions, such as a band-pass filter, a threshold function, and an inverse band-pass function. Next, (2) the learned networks are technically controllable as only a small number of driver nodes are required to move the system to a new state. Furthermore, we find that the fraction of required driver nodes is constant during evolutionary learning, suggesting a stable system design. (3), our framework allows multiplexing of different computations using the same network. For example, using a binary representation of the inputs, the network can readily compute three different input-output functions. Finally, (4) the proposed evolutionary learning demonstrates transfer learning. If the system learns one function A, then learning B requires on average less number of steps as compared to learning B from tabula rasa. We conclude that the constrained evolutionary learning produces large robust controllable circuits, capable of multiplexing and transfer learning. Our study suggests that network-based computations of steady-state functions, representing either cellular modules of cell-to-cell communication networks or internal molecular circuits communicating within a cell, could be a powerful model for biologically inspired computing. This complements conceptualizations such as attractor based models, or reservoir computing.

Deep learning models are vulnerable to external attacks. In this paper, we propose a Reinforcement Learning (RL) based approach to generate adversarial examples for the pre-trained (target) models. We assume a semi black-box setting where the only access an adversary has to the target model is the class probabilities obtained for the input queries. We train a Deep Q Network (DQN) agent which, with experience, learns to attack only a small portion of image pixels to generate non-targeted adversarial images. Initially, an agent explores an environment by sequentially modifying random sets of image pixels and observes its effect on the class probabilities. At the end of an episode, it receives a positive (negative) reward if it succeeds (fails) to alter the label of the image. Experimental results with MNIST, CIFAR-10 and Imagenet datasets demonstrate that our RL framework is able to learn an effective attack policy.

Abstract-- While learning visuomotor skills in an end-toend manner is appealing, deep neural networks are often uninterpretable and fail in surprising ways. For robotics tasks, such as autonomous driving, models that explicitly represent objects may be more robust to new scenes and provide intuitive visualizations. We describe a taxonomy of "object-centric" models which leverage both object instances and end-to-end learning. In the Grand Theft Auto V simulator, we show that object centric models outperform object-agnostic methods in scenes with other vehicles and pedestrians, even with an imperfect detector. We also demonstrate that our architectures perform well on real world environments by evaluating on the Berkeley DeepDrive Video dataset. I. INTRODUCTION End-to-end approaches to visuomotor learning are appealing in their ability to discover which features of an observed environment are most relevant for a task, and to be able to exploit large amounts of training data to discover both a policy and a codependent visual representation. Yet, the key benefit of such approaches--that they learn from task experience--is also their Achilles heel when it comes to many real-world settings, where behavioral training data is not unlimited and correct perception of "long-tail" visual phenomena can be critical for robust performance.

User interaction with voice-powered agents generates large amounts of unlabeled utterances. In this paper, we explore techniques to efficiently transfer the knowledge from these unlabeled utterances to improve model performance on Spoken Language Understanding (SLU) tasks. We use Embeddings from Language Model (ELMo) to take advantage of unlabeled data by learning contextualized word representations. Additionally, we propose ELMo-Light (ELMoL), a faster and simpler unsupervised pre-training method for SLU. Our findings suggest unsupervised pre-training on a large corpora of unlabeled utterances leads to significantly better SLU performance compared to training from scratch and it can even outperform conventional supervised transfer. Additionally, we show that the gains from unsupervised transfer techniques can be further improved by supervised transfer. The improvements are more pronounced in low resource settings and when using only 1000 labeled in-domain samples, our techniques match the performance of training from scratch on 10-15x more labeled in-domain data.