Radiology reports are the main form of communication between radiologists and other clinicians, and contain important information for patient care. However in order to use this information for research it is necessary to convert the raw text into structured data suitable for analysis. Domain specific contextual word embeddings have been shown to achieve impressive accuracy at such natural language processing tasks in medicine. In this work we pre-trained a contextual embedding BERT model using breast radiology reports and developed a classifier that incorporated the embedding with auxiliary global textual features in order to perform a section tokenization task. This model achieved a 98% accuracy at segregating free text reports into sections of information outlined in the Breast Imaging Reporting and Data System (BI-RADS) lexicon, a significant improvement over the Classic BERT model without auxiliary information. We then evaluated whether using section tokenization improved the downstream extraction of the following fields: modality/procedure, previous cancer, menopausal status, purpose of exam, breast density and background parenchymal enhancement. Using the BERT model pre-trained on breast radiology reports combined with section tokenization resulted in an overall accuracy of 95.9% in field extraction. This is a 17% improvement compared to an overall accuracy of 78.9% for field extraction for models without section tokenization and with Classic BERT embeddings. Our work shows the strength of using BERT in radiology report analysis and the advantages of section tokenization in identifying key features of patient factors recorded in breast radiology reports.

A classical computer works with ones and zeros, whereas a quantum computer uses ones, zeros, and superpositions of ones and zeros, which enables quantum computers to perform a vast number of calculations simultaneously compared to classical computers. In a cloud-supported cyber-physical system environment, running a machine learning application in quantum computers is often difficult, due to the existing limitations of the current quantum devices. However, with the combination of quantum-classical neural networks (NN), complex and high-dimensional features can be extracted by the classical NN to a reduced but more informative feature space to be processed by the existing quantum computers. In this study, we develop a hybrid quantum-classical NN to detect an amplitude shift cyber-attack on an in-vehicle control area network (CAN) dataset. We show that using the hybrid quantum classical NN, it is possible to achieve an attack detection accuracy of 94%, which is higher than a Long short-term memory (LSTM) NN (87%) or quantum NN alone (62%)

In recent years, unsupervised domain adaptation (UDA) for semantic segmentation has brought many researchers'attention. Many of them take an approach to design a complex system so as to better align the gap between source and target domain. Instead, we focus on the very basic structure of the deep neural network, Batch Normalization, and propose to replace the Sharing Affine Transformation with our proposed Separate Affine Transformation (SEAT). The proposed SEAT is simple, easily implemented and easy to integrate into existing adversarial learning based UDA methods. Also, to further improve the adaptation quality, we introduce multi level adaptation by adding the lower-level features to the higher-level ones before feeding them to the discriminator, without adding extra discriminator like others. Experiments show that the proposed methods is less complex without losing performance accuracy when compared with other UDA methods.

Plug-and-play functionality allows deep learning models to adapt well to different tasks without requiring any parameters modified. Recently, prefix-tuning was shown to be a plug-and-play method on various text generation tasks by simply inserting corresponding continuous vectors into the inputs. However, sequence labeling tasks invalidate existing plug-and-play methods since different label sets demand changes to the architecture of the model classifier. In this work, we propose the use of label word prediction instead of classification to totally reuse the architecture of pre-trained models for sequence labeling tasks. Specifically, for each task, a label word set is first constructed by selecting a high-frequency word for each class respectively, and then, task-specific vectors are inserted into the inputs and optimized to manipulate the model predictions towards the corresponding label words. As a result, by simply switching the plugin vectors on the input, a frozen pre-trained language model is allowed to perform different tasks. Experimental results on three sequence labeling tasks show that the performance of the proposed method can achieve comparable performance with standard fine-tuning with only 0.1\% task-specific parameters. In addition, our method is up to 70 times faster than non-plug-and-play methods while switching different tasks under the resource-constrained scenario.

Pre-trained language models (PLMs), such as BERT and GPT, have revolutionized the field of NLP, not only in the general domain but also in the biomedical domain. Most prior efforts in building biomedical PLMs have resorted simply to domain adaptation and focused mainly on English. In this work we introduce eHealth, a biomedical PLM in Chinese built with a new pre-training framework. This new framework trains eHealth as a discriminator through both token-level and sequence-level discrimination. The former is to detect input tokens corrupted by a generator and select their original signals from plausible candidates, while the latter is to further distinguish corruptions of a same original sequence from those of the others. As such, eHealth can learn language semantics at both the token and sequence levels. Extensive experiments on 11 Chinese biomedical language understanding tasks of various forms verify the effectiveness and superiority of our approach. The pre-trained model is available to the public at \url{https://github.com/PaddlePaddle/Research/tree/master/KG/eHealth} and the code will also be released later.

In this paper, we introduce a variational Bayesian algorithm (VBA) for image blind deconvolution. Our generic framework incorporates smoothness priors on the unknown blur/image and possible affine constraints (e.g., sum to one) on the blur kernel. One of our main contributions is the integration of VBA within a neural network paradigm, following an unrolling methodology. The proposed architecture is trained in a supervised fashion, which allows us to optimally set two key hyperparameters of the VBA model and lead to further improvements in terms of resulting visual quality. Various experiments involving grayscale/color images and diverse kernel shapes, are performed. The numerical examples illustrate the high performance of our approach when compared to state-of-the-art techniques based on optimization, Bayesian estimation, or deep learning.

Nowadays, using vibration data in conjunction with pattern recognition methods is one of the most common fault detection strategies for structures. However, their performances depend on the features extracted from vibration data, the features selected to train the classifier, and the classifier used for pattern recognition. Deep learning facilitates the fault detection procedure by automating the feature extraction and selection, and classification procedure. Though, deep learning approaches have challenges in designing its structure and tuning its hyperparameters, which may result in a low generalization capability. Therefore, this study proposes an ensemble deep learning framework based on a convolutional neural network (CNN) and Dempster-Shafer theory (DST), called CNN-DST. In this framework, several CNNs with the proposed structure are first trained, and then, the outputs of the CNNs selected by the proposed technique are combined by using an improved DST-based method. To validate the proposed CNN-DST framework, it is applied to an experimental dataset created by the broadband vibrational responses of polycrystalline Nickel alloy first-stage turbine blades with different types and severities of damage. Through statistical analysis, it is shown that the proposed CNN-DST framework classifies the turbine blades with an average prediction accuracy of 97.19%. The proposed CNN-DST framework is benchmarked with other state-of-the-art classification methods, demonstrating its high performance. The robustness of the proposed CNN-DST framework with respect to measurement noise is investigated, showing its high noise-resistance. Further, bandwidth analysis reveals that most of the required information for detecting faulty samples is available in a small frequency range.

This paper presents a novel open-domain dialogue generation model emphasizing the differentiation of speakers in multi-turn conversations. Differing from prior work that solely relies on the content of conversation history to generate a response, we argue that capturing relative social relations among utterances (i.e., generated by either the same speaker or different persons) benefits the machine capturing fine-grained context information from a conversation history to improve context coherence in the generated response. Given that, we propose a speaker-aware Parallel Hierarchical Attentive Encoder-Decoder (PHAED) model that aims to model each utterance with the awareness of its speaker and contextual associations with the same speaker's previous messages. Specifically, in a conversation involving two speakers, we regard the utterances from one speaker as responses and those from the other as queries. After understanding queries via our encoder with inner-query and inter-query encodings, our decoder reuses the hidden states of previously generated responses, instead of reconstructing these by the encoder, to generate a new response. Our empirical results show that PHAED outperforms the state-of-the-art in both automatic and human evaluations. Furthermore, our ablation study shows that dialogue models with speaker tokens can generally decrease the possibility of generating non-coherent responses regarding the conversation context.

Numerous recent works utilize bi-Lipschitz regularization of neural network layers to preserve relative distances between data instances in the feature spaces of each layer. In previous works, features extracted with a distance sensitive model are used to construct feature covariance matrices which are used in deterministic uncertainty estimation or OOD detection. However, in cases where there is a distribution over tasks, these methods result in covariances which are sub-optimal, as they may not leverage all of the meta information which can be shared among tasks. With the use of an attentive set encoder, we propose to meta learn either diagonal or diagonal plus low-rank factors to efficiently construct task specific covariance matrices. Additionally, we propose an inference procedure which utilizes scaled energy to achieve a final predictive distribution which can better separate OOD data, and is well calibrated under a distributional dataset shift. Accurate uncertainty in predictions (calibration) lies at the heart of being able to trust decisions made by deep neural networks (DNNs).

Messenger RNA-based medicines hold immense potential, as evidenced by their rapid deployment as COVID-19 vaccines. However, worldwide distribution of mRNA molecules has been limited by their thermostability, which is fundamentally limited by the intrinsic instability of RNA molecules to a chemical degradation reaction called in-line hydrolysis. Predicting the degradation of an RNA molecule is a key task in designing more stable RNA-based therapeutics. Here, we describe a crowdsourced machine learning competition ("Stanford OpenVaccine") on Kaggle, involving single-nucleotide resolution measurements on 6043 102-130-nucleotide diverse RNA constructs that were themselves solicited through crowdsourcing on the RNA design platform Eterna. The entire experiment was completed in less than 6 months. Winning models demonstrated test set errors that were better by 50% than the previous state-of-the-art DegScore model. Furthermore, these models generalized to blindly predicting orthogonal degradation data on much longer mRNA molecules (504-1588 nucleotides) with improved accuracy over DegScore and other models. Top teams integrated natural language processing architectures and data augmentation techniques with predictions from previous dynamic programming models for RNA secondary structure. These results indicate that such models are capable of representing in-line hydrolysis with excellent accuracy, supporting their use for designing stabilized messenger RNAs. The integration of two crowdsourcing platforms, one for data set creation and another for machine learning, may be fruitful for other urgent problems that demand scientific discovery on rapid timescales.