Transformers have become methods of choice in many applications thanks to their ability to represent complex interaction between elements. However, extending the Transformer architecture to non-sequential data such as molecules and enabling its training on small datasets remain a challenge. In this work, we introduce a Transformer-based architecture for molecule property prediction, which is able to capture the geometry of the molecule. We modify the classical positional encoder by an initial encoding of the molecule geometry, as well as a learned gated self-attention mechanism. We further suggest an augmentation scheme for molecular data capable of avoiding the overfitting induced by the overparameterized architecture. The proposed framework outperforms the state-of-the-art methods while being based on pure machine learning solely, i.e. the method does not incorporate domain knowledge from quantum chemistry and does not use extended geometric inputs beside the pairwise atomic distances.

Most current affect scales and sentiment analysis on written text focus on quantifying valence (sentiment) -- the most primary dimension of emotion. However, emotions are broader and more complex than valence. Distinguishing negative emotions of similar valence could be important in contexts such as mental health. This project proposes a semi-supervised machine learning model (DASentimental) to extract depression, anxiety and stress from written text. First, we trained the model to spot how sequences of recalled emotion words by $N=200$ individuals correlated with their responses to the Depression Anxiety Stress Scale (DASS-21). Within the framework of cognitive network science, we model every list of recalled emotions as a walk over a networked mental representation of semantic memory, with emotions connected according to free associations in people's memory. Among several tested machine learning approaches, we find that a multilayer perceptron neural network trained on word sequences and semantic network distances can achieve state-of-art, cross-validated predictions for depression ($R = 0.7$), anxiety ($R = 0.44$) and stress ($R = 0.52$). Though limited by sample size, this first-of-its-kind approach enables quantitative explorations of key semantic dimensions behind DAS levels. We find that semantic distances between recalled emotions and the dyad "sad-happy" are crucial features for estimating depression levels but are less important for anxiety and stress. We also find that semantic distance of recalls from "fear" can boost the prediction of anxiety but it becomes redundant when the "sad-happy" dyad is considered. Adopting DASentimental as a semi-supervised learning tool to estimate DAS in text, we apply it to a dataset of 142 suicide notes. We conclude by discussing key directions for future research enabled by artificial intelligence detecting stress, anxiety and depression.

Intent classification (IC) and slot filling (SF) are critical building blocks in task-oriented dialogue systems. These two tasks are closely-related and can flourish each other. Since only a few utterances can be utilized for identifying fast-emerging new intents and slots, data scarcity issue often occurs when implementing IC and SF. However, few IC/SF models perform well when the number of training samples per class is quite small. In this paper, we propose a novel explicit-joint and supervised-contrastive learning framework for few-shot intent classification and slot filling. Its highlights are as follows. (i) The model extracts intent and slot representations via bidirectional interactions, and extends prototypical network to achieve explicit-joint learning, which guarantees that IC and SF tasks can mutually reinforce each other. (ii) The model integrates with supervised contrastive learning, which ensures that samples from same class are pulled together and samples from different classes are pushed apart. In addition, the model follows a not common but practical way to construct the episode, which gets rid of the traditional setting with fixed way and shot, and allows for unbalanced datasets. Extensive experiments on three public datasets show that our model can achieve promising performance.

Constructing large-scaled medical knowledge graphs can significantly boost healthcare applications for medical surveillance, bring much attention from recent research. An essential step in constructing large-scale MKG is extracting information from medical reports. Recently, information extraction techniques have been proposed and show promising performance in biomedical information extraction. However, these methods only consider limited types of entity and relation due to the noisy biomedical text data with complex entity correlations. Thus, they fail to provide enough information for constructing MKGs and restrict the downstream applications. To address this issue, we propose Biomedical Information Extraction, a hybrid neural network to extract relations from biomedical text and unstructured medical reports. Our model utilizes a multi-head attention enhanced graph convolutional network to capture the complex relations and context information while resisting the noise from the data. We evaluate our model on two major biomedical relationship extraction tasks, chemical-disease relation and chemical-protein interaction, and a cross-hospital pan-cancer pathology report corpus. The results show that our method achieves superior performance than baselines. Furthermore, we evaluate the applicability of our method under a transfer learning setting and show that BioIE achieves promising performance in processing medical text from different formats and writing styles.

Personalized diagnoses have not been possible due to sear amount of data pathologists have to bear during the day-to-day routine. This lead to the current generalized standards that are being continuously updated as new findings are reported. It is noticeable that these effective standards are developed based on a multi-source heterogeneous data, including whole-slide images and pathology and clinical reports. In this study, we propose a framework that combines pathological images and medical reports to generate a personalized diagnosis result for individual patient. We use nuclei-level image feature similarity and content-based deep learning method to search for a personalized group of population with similar pathological characteristics, extract structured prognostic information from descriptive pathology reports of the similar patient population, and assign importance of different prognostic factors to generate a personalized pathological diagnosis result. We use multi-source heterogeneous data from TCGA (The Cancer Genome Atlas) database. The result demonstrate that our framework matches the performance of pathologists in the diagnosis of renal cell carcinoma. This framework is designed to be generic, thus could be applied for other types of cancer. The weights could provide insights to the known prognostic factors and further guide more precise clinical treatment protocols.

Recent development in brain-machine interface technology has made seizure prediction possible. However, the communication of large volume of electrophysiological signals between sensors and processing apparatus and related computation become two major bottlenecks for seizure prediction systems due to the constrained bandwidth and limited computation resource, especially for wearable and implantable medical devices. Although compressive sensing (CS) can be adopted to compress the signals to reduce communication bandwidth requirement, it needs a complex reconstruction procedure before the signal can be used for seizure prediction. In this paper, we propose C$^2$SP-Net, to jointly solve compression, prediction, and reconstruction with a single neural network. A plug-and-play in-sensor compression matrix is constructed to reduce transmission bandwidth requirement. The compressed signal can be used for seizure prediction without additional reconstruction steps. Reconstruction of the original signal can also be carried out in high fidelity. Prediction accuracy, sensitivity, false prediction rate, and reconstruction quality of the proposed framework are evaluated under various compression ratios. The experimental results illustrate that our model outperforms the competitive state-of-the-art baselines by a large margin in prediction accuracy. In particular, our proposed method produces an average loss of 0.35 % in prediction accuracy with a compression ratio ranging from 1/2 to 1/16.

The question how neural systems (of humans) can perform reasoning is still far from being solved. We posit that the process of forming Concepts is a fundamental step required for this. We argue that, first, Concepts are formed as closed representations, which are then consolidated by relating them to each other. Here we present a model system (agent) with a small neural network that uses realistic learning rules and receives only feedback from the environment in which the agent performs virtual actions. First, the actions of the agent are reflexive. In the process of learning, statistical regularities in the input lead to the formation of neuronal pools representing relations between the entities observed by the agent from its artificial world. This information then influences the behavior of the agent via feedback connections replacing the initial reflex by an action driven by these relational representations. We hypothesize that the neuronal pools representing relational information can be considered as primordial Concepts, which may in a similar way be present in some pre-linguistic animals, too. We argue that systems such as this can help formalizing the discussion about what constitutes Concepts and serve as a starting point for constructing artificial cogitating systems.

This paper focuses on EEG-based visual recognition, aiming to predict the visual object class observed by a subject based on his/her EEG signals. One of the main challenges is the large variation between signals from different subjects. It limits recognition systems to work only for the subjects involved in model training, which is undesirable for real-world scenarios where new subjects are frequently added. This limitation can be alleviated by collecting a large amount of data for each new user, yet it is costly and sometimes infeasible. To make the task more practical, we introduce a novel problem setting, namely subject adaptive EEG-based visual recognition. In this setting, a bunch of pre-recorded data of existing users (source) is available, while only a little training data from a new user (target) are provided. At inference time, the model is evaluated solely on the signals from the target user. This setting is challenging, especially because training samples from source subjects may not be helpful when evaluating the model on the data from the target subject. To tackle the new problem, we design a simple yet effective baseline that minimizes the discrepancy between feature distributions from different subjects, which allows the model to extract subject-independent features. Consequently, our model can learn the common knowledge shared among subjects, thereby significantly improving the recognition performance for the target subject. In the experiments, we demonstrate the effectiveness of our method under various settings. Our code is available at https://github.com/DeepBCI/Deep-BCI/tree/master/1_Intelligent_BCI/Subject_Adaptive_EEG_based_Visual_Recognition.

It is estimated that about 40%-50% of total electricity consumption in commercial buildings can be attributed to Heating, Ventilation, and Air Conditioning (HVAC) systems. Minimizing the energy cost while considering the thermal comfort of the occupants is very challenging due to unknown and complex relationships between various HVAC controls and thermal dynamics inside a building. To this end, we present a multi-agent, distributed deep reinforcement learning (DRL) framework based on Energy Plus simulation environment for optimizing HVAC in commercial buildings. This framework learns the complex thermal dynamics in the building and takes advantage of the differential effect of cooling and heating systems in the building to reduce energy costs, while maintaining the thermal comfort of the occupants. With adaptive penalty, the RL algorithm can be prioritized for energy savings or maintaining thermal comfort. Using DRL, we achieve more than 75\% savings in energy consumption. The distributed DRL framework can be scaled to multiple GPUs and CPUs of heterogeneous types.

With the development of the Internet, ways of obtaining important data such as passwords and logins or sensitive personal data have increased. One of the ways to extract such information is page impersonation, also called phishing. Such websites do not provide service but collect sensitive details from the user. Here, we present you with ways to detect such malicious URLs with state of art accuracy with neural networks. Different from previous works, where web content, URL or traffic statistics are examined, we analyse only the URL text, making it faster and which detects zero-day attacks. The network is optimised and can be used even on small devices such as Ras-Pi without a change in performance.