Topographic representation of visually evoked emotional experiences in the human cerebral cortex.

Affective neuroscience seeks to uncover the neural underpinnings of emotions that humans experience. However, it remains unclear whether an affective space underlies the discrete emotion categories in the human brain, and how it relates to the hypothesized affective dimensions. To address this question, we developed a voxel-wise encoding model to investigate the cortical organization of human emotions. Results revealed that the distributed emotion representations are constructed through a fundamental affective space. We further compared each dimension of this space to 14 hypothesized affective dimensions, and found that many affective dimensions are captured by the fundamental affective space. Our results suggest that emotional experiences are represented by broadly spatial overlapping cortical patterns and form smooth gradients across large areas of the cortex. This finding reveals the specific structure of the affective space and its relationship to hypothesized affective dimensions, while highlighting the distributed nature of emotional representations in the cortex.

Decoding Visual Neural Representations by Multimodal Learning of Brain-Visual-Linguistic Features.

Decoding human visual neural representations is a challenging task with great scientific significance in revealing vision-processing mechanisms and developing brain-like intelligent machines. Most existing methods are difficult to generalize to novel categories that have no corresponding neural data for training. The two main reasons are 1) the under-exploitation of the multimodal semantic knowledge underlying the neural data and 2) the small number of paired (stimuli-responses) training data. To overcome these limitations, this paper presents a generic neural decoding method called BraVL that uses multimodal learning of brain-visual-linguistic features. We focus on modeling the relationships between brain, visual and linguistic features via multimodal deep generative models. Specifically, we leverage the mixture-of-product-of-experts formulation to infer a latent code that enables a coherent joint generation of all three modalities. To learn a more consistent joint representation and improve the data efficiency in the case of limited brain activity data, we exploit both intra- and inter-modality mutual information maximization regularization terms. In particular, our BraVL model can be trained under various semi-supervised scenarios to incorporate the visual and textual features obtained from the extra categories. Finally, we construct three trimodal matching datasets, and the extensive experiments lead to some interesting conclusions and cognitive insights: 1) decoding novel visual categories from human brain activity is practically possible with good accuracy; 2) decoding models using the combination of visual and linguistic features perform much better than those using either of them alone; 3) visual perception may be accompanied by linguistic influences to represent the semantics of visual stimuli.

Graph-Enhanced Emotion Neural Decoding.

Brain signal-based emotion recognition has recently attracted considerable attention since it has powerful potential to be applied in human-computer interaction. To realize the emotional interaction of intelligent systems with humans, researchers have made efforts to decode human emotions from brain imaging data. The majority of current efforts use emotion similarities (e.g., emotion graphs) or brain region similarities (e.g., brain networks) to learn emotion and brain representations. However, the relationships between emotions and brain regions are not explicitly incorporated into the representation learning process. As a result, the learned representations may not be informative enough to benefit specific tasks, e.g., emotion decoding. In this work, we propose a novel idea of graph-enhanced emotion neural decoding, which takes advantage of a bipartite graph structure to integrate the relationships between emotions and brain regions into the neural decoding process, thus helping learn better representations. Theoretical analyses conclude that the suggested emotion-brain bipartite graph inherits and generalizes the conventional emotion graphs and brain networks. Comprehensive experiments on visually evoked emotion datasets demonstrate the effectiveness and superiority of our approach.

Multi-View Multi-Label Fine-Grained Emotion Decoding From Human Brain Activity.

Decoding emotional states from human brain activity play an important role in the brain-computer interfaces. Existing emotion decoding methods still have two main limitations: one is only decoding a single emotion category from a brain activity pattern and the decoded emotion categories are coarse-grained, which is inconsistent with the complex emotional expression of humans; the other is ignoring the discrepancy of emotion expression between the left and right hemispheres of the human brain. In this article, we propose a novel multi-view multi-label hybrid model for fine-grained emotion decoding (up to 80 emotion categories) which can learn the expressive neural representations and predict multiple emotional states simultaneously. Specifically, the generative component of our hybrid model is parameterized by a multi-view variational autoencoder, in which we regard the brain activity of left and right hemispheres and their difference as three distinct views and use the product of expert mechanism in its inference network. The discriminative component of our hybrid model is implemented by a multi-label classification network with an asymmetric focal loss. For more accurate emotion decoding, we first adopt a label-aware module for emotion-specific neural representation learning and then model the dependency of emotional states by a masked self-attention mechanism. Extensive experiments on two visually evoked emotional datasets show the superiority of our method.

GREN: Graph-Regularized Embedding Network for Weakly-Supervised Disease Localization in X-Ray Images.

Locating diseases in chest X-ray images with few careful annotations saves large human effort. Recent works approached this task with innovative weakly-supervised algorithms such as multi-instance learning (MIL) and class activation maps (CAM), however, these methods often yield inaccurate or incomplete regions. One of the reasons is the neglection of the pathological implications hidden in the relationship across anatomical regions within each image and the relationship across images. In this paper, we argue that the cross-region and cross-image relationship, as contextual and compensating information, is vital to obtain more consistent and integral regions. To model the relationship, we propose the Graph Regularized Embedding Network (GREN), which leverages the intra-image and inter-image information to locate diseases on chest X-ray images. GREN uses a pre-trained U-Net to segment the lung lobes, and then models the intra-image relationship between the lung lobes using an intra-image graph to compare different regions. Meanwhile, the relationship between in-batch images is modeled by an inter-image graph to compare multiple images. This process mimics the training and decision-making process of a radiologist: comparing multiple regions and images for diagnosis. In order for the deep embedding layers of the neural network to retain structural information (important in the localization task), we use the Hash coding and Hamming distance to compute the graphs, which are used as regularizers to facilitate training. By means of this, our approach achieves the state-of-the-art result on NIH chest X-ray dataset for weakly-supervised disease localization. Our codes are accessible online.

Structured Neural Decoding With Multitask Transfer Learning of Deep Neural Network Representations.

The reconstruction of visual information from human brain activity is a very important research topic in brain decoding. Existing methods ignore the structural information underlying the brain activities and the visual features, which severely limits their performance and interpretability. Here, we propose a hierarchically structured neural decoding framework by using multitask transfer learning of deep neural network (DNN) representations and a matrix-variate Gaussian prior. Our framework consists of two stages, Voxel2Unit and Unit2Pixel. In Voxel2Unit, we decode the functional magnetic resonance imaging (fMRI) data to the intermediate features of a pretrained convolutional neural network (CNN). In Unit2Pixel, we further invert the predicted CNN features back to the visual images. Matrix-variate Gaussian prior allows us to take into account the structures between feature dimensions and between regression tasks, which are useful for improving decoding effectiveness and interpretability. This is in contrast with the existing single-output regression models that usually ignore these structures. We conduct extensive experiments on two real-world fMRI data sets, and the results show that our method can predict CNN features more accurately and reconstruct the perceived natural images and faces with higher quality.

Dynamical Channel Pruning by Conditional Accuracy Change for Deep Neural Networks.

Channel pruning is an effective technique that has been widely applied to deep neural network compression. However, many existing methods prune from a pretrained model, thus resulting in repetitious pruning and fine-tuning processes. In this article, we propose a dynamical channel pruning method, which prunes unimportant channels at the early stage of training. Rather than utilizing some indirect criteria (e.g., weight norm, absolute weight sum, and reconstruction error) to guide connection or channel pruning, we design criteria directly related to the final accuracy of a network to evaluate the importance of each channel. Specifically, a channelwise gate is designed to randomly enable or disable each channel so that the conditional accuracy changes (CACs) can be estimated under the condition of each channel disabled. Practically, we construct two effective and efficient criteria to dynamically estimate CAC at each iteration of training; thus, unimportant channels can be gradually pruned during the training process. Finally, extensive experiments on multiple data sets (i.e., ImageNet, CIFAR, and MNIST) with various networks (i.e., ResNet, VGG, and MLP) demonstrate that the proposed method effectively reduces the parameters and computations of baseline network while yielding the higher or competitive accuracy. Interestingly, if we Double the initial Channels and then Prune Half (DCPH) of them to baseline's counterpart, it can enjoy a remarkable performance improvement by shaping a more desirable structure.

Reconstructing Perceived Images From Human Brain Activities With Bayesian Deep Multiview Learning.

Neural decoding, which aims to predict external visual stimuli information from evoked brain activities, plays an important role in understanding human visual system. Many existing methods are based on linear models, and most of them only focus on either the brain activity pattern classification or visual stimuli identification. Accurate reconstruction of the perceived images from the measured human brain activities still remains challenging. In this paper, we propose a novel deep generative multiview model for the accurate visual image reconstruction from the human brain activities measured by functional magnetic resonance imaging (fMRI). Specifically, we model the statistical relationships between the two views (i.e., the visual stimuli and the evoked fMRI) by using two view-specific generators with a shared latent space. On the one hand, we adopt a deep neural network architecture for visual image generation, which mimics the stages of human visual processing. On the other hand, we design a sparse Bayesian linear model for fMRI activity generation, which can effectively capture voxel correlations, suppress data noise, and avoid overfitting. Furthermore, we devise an efficient mean-field variational inference method to train the proposed model. The proposed method can accurately reconstruct visual images via Bayesian inference. In particular, we exploit a posterior regularization technique in the Bayesian inference to regularize the model posterior. The quantitative and qualitative evaluations conducted on multiple fMRI data sets demonstrate the proposed method can reconstruct visual images more accurately than the state of the art.

Improving EEG-Based Motor Imagery Classification via Spatial and Temporal Recurrent Neural Networks.

Motor imagery (MI) based Brain-Computer Interface (BCI) is an important active BCI paradigm for recognizing movement intention of severely disabled persons. There are extensive studies about MI-based intention recognition, most of which heavily rely on staged handcrafted EEG feature extraction and classifier design. For end-to-end deep learning methods, researchers encode spatial information with convolution neural networks (CNNs) from raw EEG data. Compared with CNNs, recurrent neural networks (RNNs) allow for long-range lateral interactions between features. In this paper, we proposed a pure RNNs-based parallel method for encoding spatial and temporal sequential raw data with bidirectional Long Short- Term Memory (bi-LSTM) and standard LSTM, respectively. Firstly, we rearranged the index of EEG electrodes considering their spatial location relationship. Secondly, we applied sliding window method over raw EEG data to obtain more samples and split them into training and testing sets according to their original trial index. Thirdly, we utilized the samples and their transposed matrix as input to the proposed pure RNNs- based parallel method, which encodes spatial and temporal information simultaneously. Finally, the proposed method was evaluated in the public MI-based eegmmidb dataset and compared with the other three methods (CSP+LDA, FBCSP+LDA, and CNN-RNN method). The experiment results demonstrated the superior performance of our proposed pure RNNs-based parallel method. In the multi-class trial-wise movement intention classification scenario, our approach obtained an average accuracy of 68.20% and significantly outperformed other three methods with an 8.25% improvement of relative accuracy on average, which proves the feasibility of our approach for the real-world BCI system.