Affect is often expressed via non-verbal body language such as actions/gestures, which are vital indicators for human behaviors. Recent studies on recognition of fine-grained actions/gestures in monocular images have mainly focused on modeling spatial configuration of body parts representing body pose, human-objects interactions and variations in local appearance. The results show that this is a brittle approach since it relies on accurate body parts/objects detection. In this work, we argue that there exist local discriminative semantic regions, whose "informativeness" can be evaluated by the attention mechanism for inferring fine-grained gestures/actions. To this end, we propose a novel end-to-end \textbf{Regional Attention Network (RAN)}, which is a fully Convolutional Neural Network (CNN) to combine multiple contextual regions through attention mechanism, focusing on parts of the images that are most relevant to a given task. Our regions consist of one or more consecutive cells and are adapted from the strategies used in computing HOG (Histogram of Oriented Gradient) descriptor. The model is extensively evaluated on ten datasets belonging to 3 different scenarios: 1) head pose recognition, 2) drivers state recognition, and 3) human action and facial expression recognition. The proposed approach outperforms the state-of-the-art by a considerable margin in different metrics.

In frequency division duplexing (FDD) mode, it is necessary to send the channel state information (CSI) from user equipment to base station. The downlink CSI is essential for the massive multiple-input multiple-output (MIMO) system to acquire the potential gain. Recently, deep learning is widely adopted to massive MIMO CSI feedback task and proved to be effective compared with traditional compressed sensing methods. In this paper, a novel network named ACRNet is designed to boost the feedback performance with network aggregation and parametric RuLU activation. Moreover, valid approach to expand the network architecture in exchange of better performance is first discussed in CSI feedback task. Experiments show that ACRNet outperforms loads of previous state-of-the-art feedback networks without any extra information.

Standard convolutional neural networks assume a grid structured input is available and exploit discrete convolutions as their fundamental building blocks. This limits their applicability to many real-world applications. In this paper we propose Parametric Continuous Convolution, a new learnable operator that operates over non-grid structured data. The key idea is to exploit parameterized kernel functions that span the full continuous vector space. This generalization allows us to learn over arbitrary data structures as long as their support relationship is computable. Our experiments show significant improvement over the state-of-the-art in point cloud segmentation of indoor and outdoor scenes, and lidar motion estimation of driving scenes.

We define a notion of information that an individual sample provides to the training of a neural network, and we specialize it to measure both how much a sample informs the final weights and how much it informs the function computed by the weights. Though related, we show that these quantities have a qualitatively different behavior. We give efficient approximations of these quantities using a linearized network and demonstrate empirically that the approximation is accurate for real-world architectures, such as pre-trained ResNets. We apply these measures to several problems, such as dataset summarization, analysis of under-sampled classes, comparison of informativeness of different data sources, and detection of adversarial and corrupted examples. Our work generalizes existing frameworks but enjoys better computational properties for heavily overparametrized models, which makes it possible to apply it to real-world networks. Training a deep neural network (DNN) entails extracting information from samples in a dataset and storing it in the weights of the network, so that it may be used in future inference or prediction. But how much information does a particular sample contribute to the trained model? The answer can be used to provide strong generalization bounds (if no information is used, the network is not memorizing the sample), privacy bounds (how much information the network can leak about a particular sample), and enable better interpretation of the training process and its outcome. To determine the information content of samples, we need to define and compute information. In the classical sense, information is a property of random variables, which may be degenerate for the deterministic process of computing the output of a trained DNN in response to a given input (inference). So, even posing the problem presents some technical challenges.

As progress accelerates towards AGI, the number of people who realize the significance of each new breakthrough decreases. This is the AGI Significance Paradox. There is a very old metaphor that you can boil a frog in water without it jumping out when you gradually increase the temperature. The fable goes that the frog does not have the internal models to recognize that there is a change in the water temperature. A cold-blooded creature like the frog is thought to have its temperature regulated only by the external environment.

Neural network models suffer from the phenomenon of catastrophic forgetting: a model can drastically lose its generalization ability on a task after being trained on a new task. This usually means a new task will likely override the weights that have been learned in the past (see Figure 1), and thus degrade the model performance for the past tasks. Without fixing this problem, a single neural network will not be able to adapt itself to a continuous learning scenario, because it forgets the existing information/knowledge when it learns new things. For realistic applications of deep learning, where continual learning can be crucial, catastrophic forgetting would need to be avoided. However, there is only limited study about catastrophic forgetting and its underlying causes.

We consider the problem of generating realistic traffic scenes automatically. Existing methods typically insert actors into the scene according to a set of hand-crafted heuristics and are limited in their ability to model the true complexity and diversity of real traffic scenes, thus inducing a content gap between synthesized traffic scenes versus real ones. As a result, existing simulators lack the fidelity necessary to train and test self-driving vehicles. To address this limitation, we present SceneGen, a neural autoregressive model of traffic scenes that eschews the need for rules and heuristics. In particular, given the ego-vehicle state and a high definition map of surrounding area, SceneGen inserts actors of various classes into the scene and synthesizes their sizes, orientations, and velocities. We demonstrate on two large-scale datasets SceneGen's ability to faithfully model distributions of real traffic scenes. Moreover, we show that SceneGen coupled with sensor simulation can be used to train perception models that generalize to the real world.

However, this progress is largely driven by increased computational power, namely GPU's, and bigger data sets but not due to radically new algorithms or knowledge representations. Artificial Neural Networks and Stochastic Gradient Descent, popularized in the 80's [3], remain the fundamental building blocks for most modern AI systems. While very successful for many applications, especially in vision, the purely deep-learning based approach has significant weaknesses. For instance, CNN's struggle with same-different relations [4], fail when long-chained reasoning is needed [5], are non-decomposable, cannot easily incorporate symbolic knowledge, and are hampered by a lack of model interpretability. Many current methods essentially compute higher order statistics over basic elements such as pixels, phonemes, letters or words to process inputs but do not explicitly model the building blocks and their relations in a (de)composable and interpretable way.

With the rapid development of biomedical software and hardware, a large amount of relational data interlinking genes, proteins, chemical components, drugs, diseases, and symptoms has been collected for modern biomedical research. Many graph-based learning methods have been proposed to analyze such type of data, giving a deeper insight into the topology and knowledge behind the biomedical data, which greatly benefit to both academic research and industrial application for human healthcare. However, the main difficulty is how to handle high dimensionality and sparsity of the biomedical graphs. Recently, graph embedding methods provide an effective and efficient way to address the above issues. It converts graph-based data into a low dimensional vector space where the graph structural properties and knowledge information are well preserved. In this survey, we conduct a literature review of recent developments and trends in applying graph embedding methods for biomedical data. We also introduce important applications and tasks in the biomedical domain as well as associated public biomedical datasets.

In recent years, deep neural networks have had great success in machine learning and pattern recognition. Architecture size for a neural network contributes significantly to the success of any neural network. In this study, we optimize the selection process by investigating different search algorithms to find a neural network architecture size that yields the highest accuracy. We apply binary search on a very well-defined binary classification network search space and compare the results to those of linear search. We also propose how to relax some of the assumptions regarding the dataset so that our solution can be generalized to any binary classification problem. We report a 100-fold running time improvement over the naive linear search when we apply the binary search method to our datasets in order to find the best architecture candidate. By finding the optimal architecture size for any binary classification problem quickly, we hope that our research contributes to discovering intelligent algorithms for optimizing architecture size selection in machine learning.