A Simple Meta-learning Paradigm for Zero-shot Intent Classification with Mixture Attention Mechanism

Zero-shot intent classification is a vital and challenging task in dialogue systems, which aims to deal with numerous fast-emerging unacquainted intents without annotated training data. To obtain more satisfactory performance, the crucial points lie in two aspects: extracting better utterance features and strengthening the model generalization ability. In this paper, we propose a simple yet effective meta-learning paradigm for zero-shot intent classification. To learn better semantic representations for utterances, we introduce a new mixture attention mechanism, which encodes the pertinent word occurrence patterns by leveraging the distributional signature attention and multi-layer perceptron attention simultaneously. To strengthen the transfer ability of the model from seen classes to unseen classes, we reformulate zero-shot intent classification with a meta-learning strategy, which trains the model by simulating multiple zero-shot classification tasks on seen categories, and promotes the model generalization ability with a meta-adapting procedure on mimic unseen categories. Extensive experiments on two real-world dialogue datasets in different languages show that our model outperforms other strong baselines on both standard and generalized zero-shot intent classification tasks.

Large-Scale Multi-Document Summarization with Information Extraction and Compression

We develop an abstractive summarization framework independent of labeled data for multiple heterogeneous documents. Unlike existing multi-document summarization methods, our framework processes documents telling different stories instead of documents on the same topic. We also enhance an existing sentence fusion method with a uni-directional language model to prioritize fused sentences with higher sentence probability with the goal of increasing readability. Lastly, we construct a total of twelve dataset variations based on CNN/Daily Mail and the NewsRoom datasets, where each document group contains a large and diverse collection of documents to evaluate the performance of our model in comparison with other baseline systems. Our experiments demonstrate that our framework outperforms current state-of-the-art methods in this more generic setting.

Wasserstein Distributionally Robust Optimization via Wasserstein Barycenters

In many applications in statistics and machine learning, the availability of data samples from multiple sources has become increasingly prevalent. On the other hand, in distributionally robust optimization, we seek data-driven decisions which perform well under the most adverse distribution from a nominal distribution constructed from data samples within a certain distance of probability distributions. However, it remains unclear how to achieve such distributional robustness when data samples from multiple sources are available. In this paper, we propose constructing the nominal distribution in Wasserstein distributionally robust optimization problems through the notion of Wasserstein barycenter as an aggregation of data samples from multiple sources. Under specific choices of the loss function, the proposed formulation admits a tractable reformulation as a finite convex program, with powerful finite-sample and asymptotic guarantees. We illustrate our proposed method through concrete examples with nominal distributions of location-scatter families and distributionally robust maximum likelihood estimation.

Learning to Infer Belief Embedded Communication

In multi-agent collaboration problems with communication, an agent's ability to encode their intention and interpret other agents' strategies is critical for planning their future actions. This paper introduces a novel algorithm called Intention Embedded Communication (IEC) to mimic an agent's language learning ability. IEC contains a perception module for decoding other agents' intentions in response to their past actions. It also includes a language generation module for learning implicit grammar during communication with two or more agents. Such grammar, by construction, should be compact for efficient communication. Both modules undergo conjoint evolution - similar to an infant's babbling that enables it to learn a language of choice by trial and error. We utilised three multi-agent environments, namely predator/prey, traffic junction and level-based foraging and illustrate that such a co-evolution enables us to learn much quicker (50%) than state-of-the-art algorithms like MADDPG. Ablation studies further show that disabling the inferring belief module, communication module, and the hidden states reduces the model performance by 38%, 60% and 30%, respectively. Hence, we suggest that modelling other agents' behaviour accelerates another agent to learn grammar and develop a language to communicate efficiently. We evaluate our method on a set of cooperative scenarios and show its superior performance to other multi-agent baselines. We also demonstrate that it is essential for agents to reason about others' states and learn this ability by continuous communication.

Switch Trajectory Transformer with Distributional Value Approximation for Multi-Task Reinforcement Learning

We propose SwitchTT, a multi-task extension to Trajectory Transformer but enhanced with two striking features: (i) exploiting a sparsely activated model to reduce computation cost in multi-task offline model learning and (ii) adopting a distributional trajectory value estimator that improves policy performance, especially in sparse reward settings. These two enhancements make SwitchTT suitable for solving multi-task offline reinforcement learning problems, where model capacity is critical for absorbing the vast quantities of knowledge available in the multi-task dataset. More specifically, SwitchTT exploits switch transformer model architecture for multi-task policy learning, allowing us to improve model capacity without proportional computation cost. Also, SwitchTT approximates the distribution rather than the expectation of trajectory value, mitigating the effects of the Monte-Carlo Value estimator suffering from poor sample complexity, especially in the sparse-reward setting. We evaluate our method using the suite of ten sparse-reward tasks from the gym-mini-grid environment.We show an improvement of 10% over Trajectory Transformer across 10-task learning and obtain up to 90% increase in offline model training speed. Our results also demonstrate the advantage of the switch transformer model for absorbing expert knowledge and the importance of value distribution in evaluating the trajectory.

Survival Prediction of Brain Cancer with Incomplete Radiology, Pathology, Genomics, and Demographic Data

Integrating cross-department multi-modal data (e.g., radiological, pathological, genomic, and clinical data) is ubiquitous in brain cancer diagnosis and survival prediction. To date, such an integration is typically conducted by human physicians (and panels of experts), which can be subjective and semi-quantitative. Recent advances in multi-modal deep learning, however, have opened a door to leverage such a process to a more objective and quantitative manner. Unfortunately, the prior arts of using four modalities on brain cancer survival prediction are limited by a "complete modalities" setting (i.e., with all modalities available). Thus, there are still open questions on how to effectively predict brain cancer survival from the incomplete radiological, pathological, genomic, and demographic data (e.g., one or more modalities might not be collected for a patient). For instance, should we use both complete and incomplete data, and more importantly, how to use those data? To answer the preceding questions, we generalize the multi-modal learning on cross-department multi-modal data to a missing data setting. Our contribution is three-fold: 1) We introduce optimal multi-modal learning with missing data (MMD) pipeline with optimized hardware consumption and computational efficiency; 2) We extend multi-modal learning on radiological, pathological, genomic, and demographic data into missing data scenarios; 3) a large-scale public dataset (with 962 patients) is collected to systematically evaluate glioma tumor survival prediction using four modalities. The proposed method improved the C-index of survival prediction from 0.7624 to 0.8053.

ModDrop++: A Dynamic Filter Network with Intra-subject Co-training for Multiple Sclerosis Lesion Segmentation with Missing Modalities

Multiple Sclerosis (MS) is a chronic neuroinflammatory disease and multi-modality MRIs are routinely used to monitor MS lesions. Many automatic MS lesion segmentation models have been developed and have reached human-level performance. However, most established methods assume the MRI modalities used during training are also available during testing, which is not guaranteed in clinical practice. A training strategy termed Modality Dropout (ModDrop) has been applied to MS lesion segmentation to achieve the state-of-the-art performance for missing modality. We present a novel method dubbed ModDrop++ to train a unified network adaptive to an arbitrary number of input MRI sequences. Moreover, ModDrop++ can be easily applied to any existing model architectures. Specifically, ModDrop++ upgrades the main idea of ModDrop in two key ways. First, we devise a plug-and-play dynamic head and adopt a filter scaling strategy to improve the expressiveness of the network. Second, we design a co-training strategy to leverage the intra-subject relation between full modality and missing modality. In particular, the intra-subject co-training strategy aims to guide the dynamic head to generate similar feature representations between the full- and missing-modality data from the same subject. We use two public MS datasets to show the superiority of ModDrop++. Source code and trained models are available at https://github.com/han-liu/ModDropPlusPlus.

Modeling and Validating Temporal Rules with Semantic Petri-Net for Digital Twins

Semantic rule checking on RDFS/OWL data has been widely used in the construction industry. At present, semantic rule checking is mainly performed on static models. There are still challenges in integrating temporal models and semantic models for combined rule checking. In this paper, Semantic Petri-Net (SPN) is proposed as a novel temporal modeling and validating method, which implements the states and transitions of the Colored Petri-Net directly based on RDFS and SPARQL, and realizes two-way sharing of knowledge between domain semantic webs and temporal models in the runtime. Several cases are provided to demonstrate the possible applications in digital twins with concurrent state changes and dependencies.

Synthetic CT Skull Generation for Transcranial MR Imaging-Guided Focused Ultrasound Interventions with Conditional Adversarial Networks

Transcranial MRI-guided focused ultrasound (TcMRgFUS) is a therapeutic ultrasound method that focuses sound through the skull to a small region noninvasively under MRI guidance. It is clinically approved to thermally ablate regions of the thalamus and is being explored for other therapies, such as blood brain barrier opening and neuromodulation. To accurately target ultrasound through the skull, the transmitted waves must constructively interfere at the target region. However, heterogeneity of the sound speed, density, and ultrasound attenuation in different individuals' skulls requires patient-specific estimates of these parameters for optimal treatment planning. CT imaging is currently the gold standard for estimating acoustic properties of an individual skull during clinical procedures, but CT imaging exposes patients to radiation and increases the overall number of imaging procedures required for therapy. A method to estimate acoustic parameters in the skull without the need for CT would be desirable. Here, we synthesized CT images from routinely acquired T1-weighted MRI by using a 3D patch-based conditional generative adversarial network and evaluated the performance of synthesized CT images for treatment planning with transcranial focused ultrasound. We compared the performance of synthetic CT to real CT images using Kranion and k-Wave acoustic simulation. Our work demonstrates the feasibility of replacing real CT with the MR-synthesized CT for TcMRgFUS planning.

A Multi-rater Comparative Study of Automatic Target Localization Methods for Epilepsy Deep Brain Stimulation Procedures

Epilepsy is the fourth most common neurological disorder and affects people of all ages worldwide. Deep Brain Stimulation (DBS) has emerged as an alternative treatment option when anti-epileptic drugs or resective surgery cannot lead to satisfactory outcomes. To facilitate the planning of the procedure and for its standardization, it is desirable to develop an algorithm to automatically localize the DBS stimulation target, i.e., Anterior Nucleus of Thalamus (ANT), which is a challenging target to plan. In this work, we perform an extensive comparative study by benchmarking various localization methods for ANT-DBS. Specifically, the methods involved in this study include traditional registration method and deep-learning-based methods including heatmap matching and differentiable spatial to numerical transform (DSNT). Our experimental results show that the deep-learning (DL)-based localization methods that are trained with pseudo labels can achieve a performance that is comparable to the inter-rater and intra-rater variability and that they are orders of magnitude faster than traditional methods.