Abstract--This work investigates the use of deep learning to perform user-cell association for sum-rate maximization in Massive MIMO networks. It is shown how a deep neural network can be trained to approach the optimal association rule with a much more limited computational complexity, thus enabling to update the association rule in real-time, on the basis of the mobility patterns of users. In particular, the proposed neural network design requires as input only the users' geographical positions. Numerical results show that it guarantees the same performance of traditional optimization-oriented methods. I. INTRODUCTION 5G wireless networks are scheduled to be rolled-out in only a couple of years.

We propose a new deep learning architecture for the tasks of semantic segmentation and depth prediction from RGB-D images. We revise the state of art based on the RGB and depth feature fusion, where both modalities are assumed to be available at train and test time. We propose a new architecture where the feature fusion is replaced with a common deep representation. Combined with an encoder-decoder type of the network, the architecture can jointly learn models for semantic segmentation and depth estimation based on their common representation. This representation, inspired by multi-view learning, offers several important advantages, such as using one modality available at test time to reconstruct the missing modality. In the RGB-D case, this enables the cross-modality scenarios, such as using depth data for semantically segmentation and the RGB images for depth estimation. We demonstrate the effectiveness of the proposed network on two publicly available RGB-D datasets. The experimental results show that the proposed method works well in both semantic segmentation and depth estimation tasks.

We propose a novel data augmentation method for labeled sentences called conditional BERT contextual augmentation. Data augmentation methods are often applied to prevent overfitting and improve generalization of deep neural network models. Recently proposed contextual augmentation augments labeled sentences by randomly replacing words with more varied substitutions predicted by language model. BERT demonstrates that a deep bidirectional language model is more powerful than either an unidirectional language model or the shallow concatenation of a forward and backward model. We retrofit BERT to conditional BERT by introducing a new conditional masked language model\footnote{The term "conditional masked language model" appeared once in original BERT paper, which indicates context-conditional, is equivalent to term "masked language model". In our paper, "conditional masked language model" indicates we apply extra label-conditional constraint to the "masked language model".} task. The well trained conditional BERT can be applied to enhance contextual augmentation. Experiments on six various different text classification tasks show that our method can be easily applied to both convolutional or recurrent neural networks classifier to obtain obvious improvement.

AI today is described in breathless terms as computer algorithms that use silicon incarnations of our organic brains to learn and reason about the world, intelligent superhumans rapidly making their creators obsolete. The reality could not be further from the truth. As deep learning moves from the lab into production use in mission critical fields from medicine to driverless cars, we must recognize its very real limitations as nothing more than a pile of software code and statistics, rather than the learning and thinking intelligences we describe them as. Every day data scientists build machine learning algorithms to make sense of the world and harness large piles of data into marketable insights. As guided machine assistance tools, they operate much like the large classical observation equipment of the traditional sciences, software microscopes and telescopes onto society.

Spiking neural networks (SNNs) possess energy-efficient potential due to event-based computation. However, supervised training of SNNs remains a challenge as spike activities are non-differentiable. Previous SNNs training methods can basically be categorized into two classes, backpropagation-like training methods and plasticity-based learning methods. The former methods are dependent on energy-inefficient real-valued computation and non-local transmission, as also required in artificial neural networks (ANNs), while the latter either be considered biologically implausible or exhibit poor performance. Hence, biologically plausible (bio-plausible) high-performance supervised learning (SL) methods for SNNs remain deficient. In this paper, we proposed a novel bio-plausible SNN model for SL based on the symmetric spike-timing dependent plasticity (sym-STDP) rule found in neuroscience. By combining the sym-STDP rule with bio-plausible synaptic scaling and intrinsic plasticity of the dynamic threshold, our SNN model implemented SL well and achieved good performance in the benchmark recognition task (MNIST). To reveal the underlying mechanism of our SL model, we visualized both layer-based activities and synaptic weights using the t-distributed stochastic neighbor embedding (t-SNE) method after training and found that they were well clustered, thereby demonstrating excellent classification ability. As the learning rules were bio-plausible and based purely on local spike events, our model could be easily applied to neuromorphic hardware for online training and may be helpful for understanding SL information processing at the synaptic level in biological neural systems.

The manufacturing sector is heavily influenced by artificial intelligence-based technologies with the extraordinary increases in computational power and data volumes. It has been reported that 35% of US manufacturers are currently collecting data from sensors for manufacturing processes enhancement. Nevertheless, many are still struggling to achieve the 'Industry 4.0', which aims to achieve nearly 50% reduction in maintenance cost and total machine downtime by proper health management. For increasing productivity and reducing operating costs, a central challenge lies in the detection of faults or wearing parts in machining operations. Here we propose a data-driven, end-to-end framework for monitoring of manufacturing systems. This framework, derived from deep learning techniques, evaluates fused sensory measurements to detect and even predict faults and wearing conditions. This work exploits the predictive power of deep learning to extract hidden degradation features from noisy data. We demonstrate the proposed framework on several representative experimental manufacturing datasets drawn from a wide variety of applications, ranging from mechanical to electrical systems. Results reveal that the framework performs well in all benchmark applications examined and can be applied in diverse contexts, indicating its potential for use as a critical corner stone in smart manufacturing.

The rise of deep learning has brought artificial intelligence (AI) to the forefront. The ultimate goal of AI is to realize machines with human mind and consciousness, but existing achievements mainly simulate intelligent behavior on computer platforms. These achievements all belong to weak AI rather than strong AI. How to achieve strong AI is not known yet in the field of intelligence science. Currently, this field is calling for a new paradigm, especially Theory of Cognitive Relativity (TCR). The TCR aims to summarize a simple and elegant set of first principles about the nature of intelligence, at least including the Principle of World's Relativity and the Principle of Symbol's Relativity. The Principle of World's Relativity states that the subjective world an intelligent agent can observe is strongly constrained by the way it perceives the objective world. The Principle of Symbol's Relativity states that an intelligent agent can use any physical symbol system to describe what it observes in its subjective world. The two principles are derived from scientific facts and life experience. Thought experiments show that they are important to understand high-level intelligence and necessary to establish a scientific theory of mind and consciousness. Rather than brain-like intelligence, the TCR indeed advocates a promising change in direction to realize true AI, i.e. artificial general intelligence or artificial consciousness, particularly different from humans' and animals'. Furthermore, a TCR creed has been presented and extended to reveal the secrets of consciousness and to guide realization of conscious machines. In the sense that true AI could be diversely implemented in a brain-different way, the TCR would probably drive an intelligence revolution in combination with some additional first principles.

We consider the problem of estimating human pose and trajectory by an aerial robot with a monocular camera in near real time. We present a preliminary solution whose distinguishing feature is a dynamic classifier selection architecture. In our solution, each video frame is corrected for perspective using projective transformation. Then, two alternative feature sets are used: (i) Histogram of Oriented Gradients (HOG) of the silhouette, (ii) Convolutional Neural Network (CNN) features of the RGB image. The features (HOG or CNN) are classified using a dynamic classifier. A class is defined as a pose-viewpoint pair, and a total of 64 classes are defined to represent a forward walking and turning gait sequence. Our solution provides three main advantages: (i) Classification is efficient due to dynamic selection (4-class vs. 64-class classification). (ii) Classification errors are confined to neighbors of the true view-points. (iii) The robust temporal relationship between poses is used to resolve the left-right ambiguities of human silhouettes. Experiments conducted on both fronto-parallel videos and aerial videos confirm our solution can achieve accurate pose and trajectory estimation for both scenarios. We found using HOG features provides higher accuracy than using CNN features. For example, applying the HOG-based variant of our scheme to the 'walking on a figure 8-shaped path' dataset (1652 frames) achieved estimation accuracies of 99.6% for viewpoints and 96.2% for number of poses.

Deep convolutional neural networks have been widely used in numerous applications, but their demanding storage and computational resource requirements prevent their applications on mobile devices. Knowledge distillation aims to optimize a portable student network by taking the knowledge from a well-trained heavy teacher network. Traditional teacher-student based methods used to rely on additional fully-connected layers to bridge intermediate layers of teacher and student networks, which brings in a large number of auxiliary parameters. In contrast, this paper aims to propagate information from teacher to student without introducing new variables which need to be optimized. We regard the teacher-student paradigm from a new perspective of feature embedding. By introducing the locality preserving loss, the student network is encouraged to generate the low-dimensional features which could inherit intrinsic properties of their corresponding high-dimensional features from teacher network. The resulting portable network thus can naturally maintain the performance as that of the teacher network. Theoretical analysis is provided to justify the lower computation complexity of the proposed method. Experiments on benchmark datasets and well-trained networks suggest that the proposed algorithm is superior to state-of-the-art teacher-student learning methods in terms of computational and storage complexity.

As Computer Vision algorithms move from passive analysis of pixels to active reasoning over semantics, the breadth of information algorithms need to reason over has expanded significantly. One of the key challenges in this vein is the ability to identify the information required to make a decision, and select an action that will recover this information. We propose an reinforcement-learning approach that maintains an distribution over its internal information, thus explicitly representing the ambiguity in what it knows, and needs to know, towards achieving its goal. Potential actions are then generated according to particles sampled from this distribution. For each potential action a distribution of the expected answers is calculated, and the value of the information gained is obtained, as compared to the existing internal information. We demonstrate this approach applied to two vision-language problems that have attracted significant recent interest, visual dialogue and visual query generation. In both cases the method actively selects actions that will best reduce its internal uncertainty, and outperforms its competitors in achieving the goal of the challenge.