Metasurfaces are being widely investigated and adopted for their promising performances in manipulating optical wavefronts and their potential for integrating multi-functionalities into one flat optical device. A key challenge in metasurface design is the non-intuitive design process that produces models and patterns from specific design requirements (commonly electromagnetic responses). A complete exploration of all design spaces can produce optimal designs but is unrealistic considering the massive amount of computation power required to explore large parameter spaces. Meanwhile, machine learning techniques, especially generative adversarial networks, have proven to be an effective solution to non-intuitive design tasks. In this paper, we present a novel conditional generative network that can generate meta-atom/metasurface designs based on different performance requirements. Compared to conventional trial-and-error or iterative optimization design methods, this new methodology is capable of producing on-demand freeform designs on a one-time calculation basis. More importantly, an increased complexity of design goals doesn't introduce further complexity into the network structure or the training process, which makes this approach suitable for multi-functional device designs. Compared to previous deep learning-based metasurface approaches, our network structure is extremely robust to train and converge, and is readily expanded to many multi-functional metasurface devices, including metasurface filters, lenses and holograms.
Metasurfaces have become a promising means for manipulating optical wavefronts in flat and high-performance optical devices. Conventional metasurface device design relies on trial-and-error methods to obtain target electromagnetic (EM) response, an approach that demands significant efforts to investigate the enormous number of possible meta-atom structures. In this paper, a deep neural network approach is introduced that significantly improves on both speed and accuracy compared to techniques currently used to assemble metasurface-based devices. Our neural network approach overcomes three key challenges that have limited previous neural-network-based design schemes: input/output vector dimensional mismatch, accurate EM-wave phase prediction, as well as adaptation to 3-D dielectric structures, and can be generically applied to a wide variety of metasurface device designs across the entire electromagnetic spectrum. Using this new methodology, examples of neural networks capable of producing on-demand designs for meta-atoms, metasurface filters, and phase-change reconfigurable metasurfaces are demonstrated.
Contextual bandit algorithms have gained increasing popularity in recommender systems, because they can learn to adapt recommendations by making exploration-exploitation trade-off. Recommender systems equipped with traditional contextual bandit algorithms are usually trained with behavioral feedback (e.g., clicks) from users on items. The learning speed can be slow because behavioral feedback by nature does not carry sufficient information. As a result, extensive exploration has to be performed. To address the problem, we propose conversational recommendation in which the system occasionally asks questions to the user about her interest. We first generalize contextual bandit to leverage not only behavioral feedback (arm-level feedback), but also verbal feedback (users' interest on categories, topics, etc.). We then propose a new UCB- based algorithm, and theoretically prove that the new algorithm can indeed reduce the amount of exploration in learning. We also design several strategies for asking questions to further optimize the speed of learning. Experiments on synthetic data, Yelp data, and news recommendation data from Toutiao demonstrate the efficacy of the proposed algorithm.
In this paper, we investigate the design and implementation of machine learning (ML) based demodulation methods in the physical layer of visible light communication (VLC) systems. We build a flexible hardware prototype of an end-to-end VLC system, from which the received signals are collected as the real data. The dataset is available online, which contains eight types of modulated signals. Then, we propose three ML demodulators based on convolutional neural network (CNN), deep belief network (DBN), and adaptive boosting (AdaBoost), respectively. Specifically, the CNN based demodulator converts the modulated signals to images and recognizes the signals by the image classification. The proposed DBN based demodulator contains three restricted Boltzmann machines (RBMs) to extract the modulation features. The AdaBoost method includes a strong classifier that is constructed by the weak classifiers with the k-nearest neighbor (KNN) algorithm. These three demodulators are trained and tested by our online open dataset. Experimental results show that the demodulation accuracy of the three data-driven demodulators drops as the transmission distance increases. A higher modulation order negatively influences the accuracy for a given transmission distance. Among the three ML methods, the AdaBoost modulator achieves the best performance.
Text similarity calculation is a fundamental problem in natural language processing and related fields. In recent years, deep neural networks have been developed to perform the task and high performances have been achieved. The neural networks are usually trained with labeled data in supervised learning, and creation of labeled data is usually very costly. In this short paper, we address unsupervised learning for text similarity calculation. We propose a new method called Word Embedding based Edit Distance (WED), which incorporates word embedding into edit distance. Experiments on three benchmark datasets show WED outperforms state-of-the-art unsupervised methods including edit distance, TF-IDF based cosine, word embedding based cosine, Jaccard index, etc.
Automatic generation of paraphrases from a given sentence is an important yet challenging task in natural language processing (NLP), and plays a key role in a number of applications such as question answering, search, and dialogue. In this paper, we present a deep reinforcement learning approach to paraphrase generation. Specifically, we propose a new framework for the task, which consists of a \textit{generator} and an \textit{evaluator}, both of which are learned from data. The generator, built as a sequence-to-sequence learning model, can produce paraphrases given a sentence. The evaluator, constructed as a deep matching model, can judge whether two sentences are paraphrases of each other. The generator is first trained by deep learning and then further fine-tuned by reinforcement learning in which the reward is given by the evaluator. For the learning of the evaluator, we propose two methods based on supervised learning and inverse reinforcement learning respectively, depending on the type of available training data. Empirical study shows that the learned evaluator can guide the generator to produce more accurate paraphrases. Experimental results demonstrate the proposed models (the generators) outperform the state-of-the-art methods in paraphrase generation in both automatic evaluation and human evaluation.
From the viewpoint of physical-layer authentication, spoofing attacks can be foiled by checking channel state information (CSI). Existing CSI-based authentication algorithms mostly require a deep knowledge of the channel to deliver decent performance. In this paper, we investigate CSI-based authenticators that can spare the effort to predetermine channel properties by utilizing deep neural networks (DNNs). We first propose a convolutional neural network (CNN)-enabled authenticator that is able to extract the local features in CSI. Next, we employ the recurrent neural network (RNN) to capture the dependencies between different frequencies in CSI. In addition, we propose to use the convolutional recurrent neural network (CRNN)---a combination of the CNN and the RNN---to learn local and contextual information in CSI for user authentication. To effectively train these DNNs, one needs a large amount of labeled channel records. However, it is often expensive to label large channel observations in the presence of a spoofer. In view of this, we further study a case in which only a small part of the the channel observations are labeled. To handle it, we extend these DNNs-enabled approaches into semi-supervised ones. This extension is based on a semi-supervised learning technique that employs both the labeled and unlabeled data to train a DNN. To be specific, our semi-supervised method begins by generating pseudo labels for the unlabeled channel samples through implementing the K-means algorithm in a semi-supervised manner. Subsequently, both the labeled and pseudo labeled data are exploited to pre-train a DNN, which is then fine-tuned based on the labeled channel records.
Few-shot learning is challenging for learning algorithms that learn each task in isolation and from scratch. In contrast, meta-learning learns from many related tasks a meta-learner that can learn a new task more accurately and faster with fewer examples, where the choice of meta-learners is crucial. In this paper, we develop Meta-SGD, an SGD-like, easily trainable meta-learner that can initialize and adapt any differentiable learner in just one step, on both supervised learning and reinforcement learning. Compared to the popular meta-learner LSTM, Meta-SGD is conceptually simpler, easier to implement, and can be learned more efficiently. Compared to the latest meta-learner MAML, Meta-SGD has a much higher capacity by learning to learn not just the learner initialization, but also the learner update direction and learning rate, all in a single meta-learning process. Meta-SGD shows highly competitive performance for few-shot learning on regression, classification, and reinforcement learning.
Safe robot navigation is a fundamental research field for autonomous robots including ground mobile robots and flying robots. The primary objective of a safe robot navigation algorithm is to guide an autonomous robot from its initial position to a target or along a desired path with obstacle avoidance. With the development of information technology and sensor technology, the implementations combining robotics with sensor network are focused on in the recent researches. One of the relevant implementations is the sensor network based robot navigation. Moreover, another important navigation problem of robotics is safe area search and map building. In this report, a global collision-free path planning algorithm for ground mobile robots in dynamic environments is presented firstly. Considering the advantages of sensor network, the presented path planning algorithm is developed to a sensor network based navigation algorithm for ground mobile robots. The 2D range finder sensor network is used in the presented method to detect static and dynamic obstacles. The sensor network can guide each ground mobile robot in the detected safe area to the target. Furthermore, the presented navigation algorithm is extended into 3D environments. With the measurements of the sensor network, any flying robot in the workspace is navigated by the presented algorithm from the initial position to the target. Moreover, in this report, another navigation problem, safe area search and map building for ground mobile robot, is studied and two algorithms are presented. In the first presented method, we consider a ground mobile robot equipped with a 2D range finder sensor searching a bounded 2D area without any collision and building a complete 2D map of the area. Furthermore, the first presented map building algorithm is extended to another algorithm for 3D map building.
Building neural networks to query a knowledge base (a table) with natural language is an emerging research topic in deep learning. An executor for table querying typically requires multiple steps of execution because queries may have complicated structures. In previous studies, researchers have developed either fully distributed executors or symbolic executors for table querying. A distributed executor can be trained in an end-to-end fashion, but is weak in terms of execution efficiency and explicit interpretability. A symbolic executor is efficient in execution, but is very difficult to train especially at initial stages. In this paper, we propose to couple distributed and symbolic execution for natural language queries, where the symbolic executor is pretrained with the distributed executor's intermediate execution results in a step-by-step fashion. Experiments show that our approach significantly outperforms both distributed and symbolic executors, exhibiting high accuracy, high learning efficiency, high execution efficiency, and high interpretability.