We address the challenging task of cross-modal moment retrieval, which aims to localize a temporal segment from an untrimmed video described by a natural language query. It poses great challenges over the proper semantic alignment between vision and linguistic domains. Existing methods independently extract the features of videos and sentences and purely utilize the sentence embedding in the multi-modal fusion stage, which do not make full use of the potential of language. In this paper, we present Language Guided Networks (LGN), a new framework that leverages the sentence embedding to guide the whole process of moment retrieval. In the first feature extraction stage, we propose to jointly learn visual and language features to capture the powerful visual information which can cover the complex semantics in the sentence query. Specifically, the early modulation unit is designed to modulate the visual feature extractor's feature maps by a linguistic embedding. Then we adopt a multi-modal fusion module in the second fusion stage. Finally, to get a precise localizer, the sentence information is utilized to guide the process of predicting temporal positions. Specifically, the late guidance module is developed to linearly transform the output of localization networks via the channel attention mechanism. The experimental results on two popular datasets demonstrate the superior performance of our proposed method on moment retrieval (improving by 5.8\% in terms of Rank1@IoU0.5 on Charades-STA and 5.2\% on TACoS). The source code for the complete system will be publicly available.

Given the recent successes of Deep Learning in AI there has been increased interest in the role and need for explanations in machine learned theories. A distinct notion in this context is that of Michie's definition of Ultra-Strong Machine Learning (USML). USML is demonstrated by a measurable increase in human performance of a task following provision to the human of a symbolic machine learned theory for task performance. A recent paper demonstrates the beneficial effect of a machine learned logic theory for a classification task, yet no existing work has examined the potential harmfulness of machine's involvement in human learning. This paper investigates the explanatory effects of a machine learned theory in the context of simple two person games and proposes a framework for identifying the harmfulness of machine explanations based on the Cognitive Science literature. The approach involves a cognitive window consisting of two quantifiable bounds and it is supported by empirical evidence collected from human trials. Our quantitative and qualitative results indicate that human learning aided by a symbolic machine learned theory which satisfies a cognitive window has achieved significantly higher performance than human self learning. Results also demonstrate that human learning aided by a symbolic machine learned theory that fails to satisfy this window leads to significantly worse performance than unaided human learning.

Campaign is a frequently employed instrument in lifting up the GMV (Gross Merchandise Volume) of retailer in traditional marketing. As its counterpart in online context, digital-marketing-campaign (DMC) has being trending in recent years with the rapid development of the e-commerce. However, how to empower massive sellers on the online retailing platform the capacity of applying combined multiple digital marketing campaigns to boost their shops' revenue, is still a novel topic. In this work, a comprehensive solution of generating optimized combined multiple DMCs is presented. Firstly, a potential personalized DMC pool is generated for every retailer by a newly proposed neural network model, i.e. the DMCNet (Digital-Marketing-Campaign Net). Secondly, based on the sub-modular optimization theory and the DMC pool by DMCNet, the generated combined multiple DMCs are ranked with respect to their revenue generation strength then the top three ranked campaigns are returned to the sellers' back-end management system, so that retailers can set combined multiple DMCs for their online shops just in one-shot. Real online A/B-test shows that with the integrated solution, sellers of the online retailing platform increase their shops' GMVs with approximately 6$\%$.

We explore the use of multiple deep learning models for detecting flaws in software programs. Current, standard approaches for flaw detection rely on a single representation of a software program (e.g., source code or a program binary). We illustrate that, by using techniques from multimodal deep learning, we can simultaneously leverage multiple representations of software programs to improve flaw detection over single representation analyses. Specifically, we adapt three deep learning models from the multimodal learning literature for use in flaw detection and demonstrate how these models outperform traditional deep learning models.

We study the behavior of several black-box search algorithms used for generating adversarial examples for natural language processing (NLP) tasks. We perform a fine-grained analysis of three elements relevant to search: search algorithm, search space, and search budget. When new search methods are proposed in past work, the attack search space is often modified alongside the search method. Without ablation studies benchmarking the search algorithm change with the search space held constant, an increase in attack success rate could from an improved search method or a less restrictive search space. Additionally, many previous studies fail to properly consider the search algorithms' run-time cost, which is essential for downstream tasks like adversarial training. Our experiments provide a reproducible benchmark of search algorithms across a variety of search spaces and query budgets to guide future research in adversarial NLP. Based on our experiments, we recommend greedy attacks with word importance ranking when under a time constraint or attacking long inputs, and either beam search or particle swarm optimization otherwise. Code implementation shared via https://github.com/QData/TextAttack

Urban ride-hailing demand prediction is a crucial but challenging task for intelligent transportation system construction. Predictable ride-hailing demand can facilitate more reasonable vehicle scheduling and online car-hailing platform dispatch. Conventional deep learning methods with no external structured data can be accomplished via hybrid models of CNNs and RNNs by meshing plentiful pixel-level labeled data, but spatial data sparsity and limited learning capabilities on temporal long-term dependencies are still two striking bottlenecks. To address these limitations, we propose a new virtual graph modeling method to focus on significant demand regions and a novel Deep Multi-View Spatiotemporal Virtual Graph Neural Network (DMVST-VGNN) to strengthen learning capabilities of spatial dynamics and temporal long-term dependencies. Specifically, DMVST-VGNN integrates the structures of 1D Convolutional Neural Network, Multi Graph Attention Neural Network and Transformer layer, which correspond to short-term temporal dynamics view, spatial dynamics view and long-term temporal dynamics view respectively. In this paper, experiments are conducted on two large-scale New York City datasets in fine-grained prediction scenes. And the experimental results demonstrate effectiveness and superiority of DMVST-VGNN framework in significant citywide ride-hailing demand prediction.

Reinforcement Learning (RL) can be used to fit a mapping from patient state to a medication regimen. Prior studies have used deterministic and value-based tabular learning to learn a propofol dose from an observed anesthetic state. Deep RL replaces the table with a deep neural network and has been used to learn medication regimens from registry databases. Here we perform the first application of deep RL to closed-loop control of anesthetic dosing in a simulated environment. We use the cross-entropy method to train a deep neural network to map an observed anesthetic state to a probability of infusing a fixed propofol dosage. During testing, we implement a deterministic policy that transforms the probability of infusion to a continuous infusion rate. The model is trained and tested on simulated pharmacokinetic/pharmacodynamic models with randomized parameters to ensure robustness to patient variability. The deep RL agent significantly outperformed a proportional-integral-derivative controller (median absolute performance error 1.7% +/- 0.6 and 3.4% +/- 1.2). Modeling continuous input variables instead of a table affords more robust pattern recognition and utilizes our prior domain knowledge. Deep RL learned a smooth policy with a natural interpretation to data scientists and anesthesia care providers alike.

Relative attributes indicate the strength of a particular attribute between image pairs. We introduce a deep Siamese network with rank SVM loss function, called Deep Rank SVM (DRSVM), in order to decide which one of a pair of images has a stronger presence of a specific attribute. The network is trained in an end-to-end fashion to jointly learn the visual features and the ranking function. We demonstrate the effectiveness of our approach against the state-of-the-art methods on four image benchmark datasets: LFW-10, PubFig, UTZap50K-lexi and UTZap50K-2 datasets. DRSVM surpasses state-of-art in terms of the average accuracy across attributes, on three of the four image benchmark datasets.

We introduce VoiceFilter-Lite, a single-channel source separation model that runs on the device to preserve only the speech signals from a target user, as part of a streaming speech recognition system. Delivering such a model presents numerous challenges: It should improve the performance when the input signal consists of overlapped speech, and must not hurt the speech recognition performance under all other acoustic conditions. Besides, this model must be tiny, fast, and perform inference in a streaming fashion, in order to have minimal impact on CPU, memory, battery and latency. We propose novel techniques to meet these multi-faceted requirements, including using a new asymmetric loss, and adopting adaptive runtime suppression strength. We also show that such a model can be quantized as a 8-bit integer model and run in realtime.

The use of a few examples for each class to train a predictive model that can be generalized to novel classes is a crucial and valuable research direction in artificial intelligence. This work addresses this problem by proposing a few-shot learning (FSL) algorithm called proxy network under the architecture of meta-learning. Metric-learning based approaches assume that the data points within the same class should be close, whereas the data points in the different classes should be separated as far as possible in the embedding space. We conclude that the success of metric-learning based approaches lies in the data embedding, the representative of each class, and the distance metric. In this work, we propose a simple but effective end-to-end model that directly learns proxies for class representative and distance metric from data simultaneously. We conduct experiments on CUB and mini-ImageNet datasets in 1-shot-5-way and 5-shot-5-way scenarios, and the experimental results demonstrate the superiority of our proposed method over state-of-the-art methods. Besides, we provide a detailed analysis of our proposed method.