IJCAI 2021 论文列表

Paraphrase generation plays key roles in NLP tasks such as question answering, machine translation, and information retrieval. In this paper, we propose a novel framework for paraphrase generation. It simultaneously decodes the output sentence using a pretrained wordset-to-sequence model and a round-trip translation model. We evaluate this framework on Quora, WikiAnswers, MSCOCO and Twitter, and show its advantage over previous state-of-the-art unsupervised methods and distantly-supervised methods by significant margins on all datasets. For Quora and WikiAnswers, our framework even performs better than some strongly supervised methods with domain adaptation. Further, we show that the generated paraphrases can be used to augment the training data for machine translation to achieve substantial improvements.

Music-to-dance translation is an emerging and powerful feature in recent role-playing games. Previous works of this topic consider music-to-dance as a supervised motion generation problem based on time-series data. However, these methods require a large amount of training data pairs and may suffer from the degradation of movements. This paper provides a new solution to this task where we re-formulate the translation as a piece-wise dance phrase retrieval problem based on the choreography theory. With such a design, players are allowed to optionally edit the dance movements on top of our generation while other regression-based methods ignore such user interactivity. Considering that the dance motion capture is expensive that requires the assistance of professional dancers, we train our method under a semi-supervised learning fashion with a large unlabeled music dataset (20x than our labeled one) and also introduce self-supervised pre-training to improve the training stability and generalization performance. Experimental results suggest that our method not only generalizes well over various styles of music but also succeeds in choreography for game players. Our project including the large-scale dataset and supplemental materials is available at https://github.com/FuxiCV/music-to-dance.

Pre-trained language models such as BERT have shown remarkable effectiveness in various natural language processing tasks. However, these models usually contain millions of parameters, which prevent them from the practical deployment on resource-constrained devices. Knowledge distillation, Weight pruning, and Quantization are known to be the main directions in model compression. However, compact models obtained through knowledge distillation may suffer from significant accuracy drop even for a relatively small compression ratio. On the other hand, there are only a few attempts based on quantization designed for natural language processing tasks, and they usually require manual setting on hyper-parameters. In this paper, we proposed an automatic mixed-precision quantization framework designed for BERT that can conduct quantization and pruning simultaneously. Specifically, our proposed method leverages Differentiable Neural Architecture Search to assign scale and precision for parameters in each sub-group automatically, and at the same pruning out redundant groups of parameters. Extensive evaluations on BERT downstream tasks reveal that our proposed method beats baselines by providing the same performance with much smaller model size. We also show the possibility of obtaining the extremely light-weight model by combining our solution with orthogonal methods such as DistilBERT.

In this work we present AutoBQP, a heuristic solver for binary optimization problems. It applies automatic algorithm design techniques to search for the best heuristics for a given optimization problem. Experiments show that the solver can find algorithms which perform better than or comparable to state-of-the-art methods, and can even find new best solutions for some instances of standard benchmark sets.

Machine learning on graphs has been extensively studied in both academic and industry. However, as the literature on graph learning booms with a vast number of emerging methods and techniques, it becomes increasingly difficult to manually design the optimal machine learning algorithm for different graph-related tasks. To solve this critical challenge, automated machine learning (AutoML) on graphs which combines the strength of graph machine learning and AutoML together, is gaining attention from the research community. Therefore, we comprehensively survey AutoML on graphs in this paper, primarily focusing on hyper-parameter optimization (HPO) and neural architecture search (NAS) for graph machine learning. We further overview libraries related to automated graph machine learning and in-depth discuss AutoGL, the first dedicated open-source library for AutoML on graphs. In the end, we share our insights on future research directions for automated graph machine learning. This paper is the first systematic and comprehensive review of automated machine learning on graphs to the best of our knowledge.

The reporting and the analysis of current events around the globe has expanded from professional, editor-lead journalism all the way to citizen journalism. Nowadays, politicians and other key players enjoy direct access to their audiences through social media, bypassing the filters of official cables or traditional media. However, the multiple advantages of free speech and direct communication are dimmed by the misuse of media to spread inaccurate or misleading claims. These phenomena have led to the modern incarnation of the fact-checker --- a professional whose main aim is to examine claims using available evidence and to assess their veracity. Here, we survey the available intelligent technologies that can support the human expert in the different steps of her fact-checking endeavor. These include identifying claims worth fact-checking, detecting relevant previously fact-checked claims, retrieving relevant evidence to fact-check a claim, and actually verifying a claim. In each case, we pay attention to the challenges and the potential impact on real-world fact-checking.

Online forum that gathers participants together to solve the common issues that they are facing is considered as a promising application of utilizing collective intelligence to solve complicated real-world problems. To facilitate the discussions in online forum to proceed smoothly and to build consensus efficiently, human facilitators are introduced into the system. With the increasing sophistication of online forum, human facilitators related problems such as human bias and restricted scale become critical. Therefore, it is critical to explore approaches to support human facilitators in conducting facilitation. However, most of the existing facilitation support techniques only support predefined facilitation tasks that could be defined by static rules. In this research, we aim to explore potential solutions for supporting the human facilitators to conduct facilitation in online forum. As the first step, we have proposed a case-based reasoning (CBR)-based framework that targets support facilitation by utilizing past successful facilitation experience. Currently, our work is focusing on the specific facilitation task of detecting influential user in the online forum. In the future work, we are planning to propose approaches of solving other specific facilitation tasks such as measuring the level of agreement and encouraging participants to reach a consensus.

Data-free compression raises a new challenge because the original training dataset for a pre-trained model to be compressed is not available due to privacy or transmission issues. Thus, a common approach is to compute a reconstructed training dataset before compression. The current reconstruction methods compute the reconstructed training dataset with a generator by exploiting information from the pre-trained model. However, current reconstruction methods focus on extracting more information from the pre-trained model but do not leverage network engineering. This work is the first to consider network engineering as an approach to design the reconstruction method. Specifically, we propose the AutoReCon method, which is a neural architecture search-based reconstruction method. In the proposed AutoReCon method, the generator architecture is designed automatically given the pre-trained model for reconstruction. Experimental results show that using generators discovered by the AutoRecon method always improve the performance of data-free compression.

Recently online multi-armed bandit (MAB) is growing rapidly, as novel problem settings and algorithms motivated by various practical applications are being studied, building on the top of the classic bandit problem. However, identifying the best bandit algorithm from lots of potential candidates for a given application is not only time-consuming but also relying on human expertise, which hinders the practicality of MAB. To alleviate this problem, this paper outlines an intelligent system called AutoBandit, equipped with many out-of-the-box MAB algorithms, for automatically and adaptively choosing the best with suitable hyper-parameters online. It is effective to help a growing application for continuously maximizing cumulative rewards of its whole life-cycle. With a flexible architecture and user-friendly web-based interfaces, it is very convenient for the user to integrate and monitor online bandits in a business system. At the time of publication, AutoBandit has been deployed for various industrial applications.

We propose an audio-driven talking-head method to generate photo-realistic talking-head videos from a single reference image. In this work, we tackle two key challenges: (i) producing natural head motions that match speech prosody, and (ii)} maintaining the appearance of a speaker in a large head motion while stabilizing the non-face regions. We first design a head pose predictor by modeling rigid 6D head movements with a motion-aware recurrent neural network (RNN). In this way, the predicted head poses act as the low-frequency holistic movements of a talking head, thus allowing our latter network to focus on detailed facial movement generation. To depict the entire image motions arising from audio, we exploit a keypoint based dense motion field representation. Then, we develop a motion field generator to produce the dense motion fields from input audio, head poses, and a reference image. As this keypoint based representation models the motions of facial regions, head, and backgrounds integrally, our method can better constrain the spatial and temporal consistency of the generated videos. Finally, an image generation network is employed to render photo-realistic talking-head videos from the estimated keypoint based motion fields and the input reference image. Extensive experiments demonstrate that our method produces videos with plausible head motions, synchronized facial expressions, and stable backgrounds and outperforms the state-of-the-art.

Though convolutional neural networks (CNNs) with residual and dense aggregations have obtained much attention in image denoising, they are incapable of exploiting different levels of contextual information at every convolutional unit in order to infer different levels of noise components with a single model. In this paper, to overcome this shortcoming we present a novel attention-based pyramid dilated lattice (APDL) architecture and investigate its capability for blind image denoising. The proposed framework can effectively harness the advantages of residual and dense aggregations to achieve a great trade-off between performance, parameter efficiency, and test time. It also employs a novel pyramid dilated convolution strategy to effectively capture contextual information corresponding to different noise levels through the training of a single model. Our extensive experimental investigation verifies the effectiveness and efficiency of the APDL architecture for image denoising as well as JPEG artifacts suppression tasks.

Multi-hop machine reading comprehension (MRC) task aims to enable models to answer the compound question according to the bridging information. Existing methods that use graph neural networks to represent multiple granularities such as entities and sentences in documents update all nodes synchronously, ignoring the fact that multi-hop reasoning has a certain logical order across granular levels. In this paper, we introduce an Asynchronous Multi-grained Graph Network (AMGN) for multi-hop MRC. First, we construct a multigrained graph containing entity and sentence nodes. Particularly, we use independent parameters to represent relationship groups defined according to the level of granularity. Second, an asynchronous update mechanism based on multi-grained relationships is proposed to mimic human multi-hop reading logic. Besides, we present a question reformulation mechanism to update the latent representation of the compound question with updated graph nodes. We evaluate the proposed model on the HotpotQA dataset and achieve top competitive performance in distractor setting compared with other published models. Further analysis shows that the asynchronous update mechanism can effectively form interpretable reasoning chains at different granularity levels.

Active learning tries to learn an effective model with lowest labeling cost. Most existing active learning methods work in a synchronous way, which implies that the label querying can be performed only after the model updating in each iteration. While training models is usually time-consuming, it may lead to serious latency between two queries, especially in the crowdsourcing environments where there are many online annotators working simultaneously. This will significantly decrease the labeling efficiency and strongly limit the application of active learning in real tasks. To overcome this challenge, we propose a multi-server multi-worker framework for asynchronous active learning in the distributed environment. By maintaining two shared pools of candidate queries and labeled data respectively, the servers, the workers and the annotators efficiently corporate with each other without synchronization. Moreover, diverse sampling strategies from distributed workers are incorporated to select the most useful instances for model improving. Both theoretical analysis and experimental study validate the effectiveness of the proposed approach.

Explainable AI (XAI) has been investigated for decades and, together with AI itself, has witnessed unprecedented growth in recent years. Among various approaches to XAI, argumentative models have been advocated in both the AI and social science literature, as their dialectical nature appears to match some basic desirable features of the explanation activity. In this survey we overview XAI approaches built using methods from the field of computational argumentation, leveraging its wide array of reasoning abstractions and explanation delivery methods. We overview the literature focusing on different types of explanation (intrinsic and post-hoc), different models with which argumentation-based explanations are deployed, different forms of delivery, and different argumentation frameworks they use. We also lay out a roadmap for future work.

The Shapley value is a well recognised method for dividing the value of joint effort in cooperative games. However, computing the Shapley value is known to be computationally hard, so stratified sample-based estimation is sometimes used. For this task, we provide two contributions to the state of the art. First, we derive a novel concentration inequality that is tailored to stratified Shapley value estimation using sample variance information. Second, by sequentially choosing samples to minimize our inequality, we develop a new and more efficient method of sampling to estimate the Shapley value. We evaluate our sampling method on a suite of test cooperative games, and our results demonstrate that it outperforms or is competitive with existing stratified sample-based estimation approaches to computing the Shapley value.

Multi-Agent Path Finding (MAPF) is the challenging problem of computing collision-free paths for multiple agents. Algorithms for solving MAPF can be categorized on a spectrum. At one end are (bounded-sub)optimal algorithms that can find high-quality solutions for small problems. At the other end are unbounded-suboptimal algorithms that can solve large problems but usually find low-quality solutions. In this paper, we consider a third approach that combines the best of both worlds: anytime algorithms that quickly find an initial solution using efficient MAPF algorithms from the literature, even for large problems, and that subsequently improve the solution quality to near-optimal as time progresses by replanning subgroups of agents using Large Neighborhood Search. We compare our algorithm MAPF-LNS against a range of existing work and report significant gains in scalability, runtime to the initial solution, and speed of improving the solution.

Anomaly mining is an important problem that finds numerous applications in various real world do- mains such as environmental monitoring, cybersecurity, finance, healthcare and medicine, to name a few. In this article, I focus on two areas, (1) point-cloud and (2) graph-based anomaly mining. I aim to present a broad view of each area, and discuss classes of main research problems, recent trends and future directions. I conclude with key take-aways and overarching open problems. Disclaimer. I try to provide an overview of past and recent trends in both areas within 4 pages. Undoubtedly, these are my personal view of the trends, which can be organized differently. For brevity, I omit all technical details and refer to corresponding papers. Again, due to space limit, it is not possible to include all (even most relevant) references, but a few representative examples.

This paper presents a survey of researches in analogical reasoning whose building block are analogical proportions which are statements of the form “a is to b as c is to d”. They have been developed in the last twenty years within an Artificial Intelligence perspective. After discussing their formal modeling with the associated inference mechanism, the paper reports the main results obtained in various AI domains ranging from computational linguistics to classification, including image processing, I.Q. tests, case based reasoning, preference learning, and formal concepts analysis. The last section discusses some new theoretical concerns, and the potential of analogical proportions in other areas such as argumentation, transfer learning, and XAI.

While the knowledge about the structures of a system’s underlying causal relationships is crucial within many real-world scenarios, the omnipresence of heterogeneous data characteristics impedes applying methods for causal structure learning (CSL). In this dissertation project, we reduce the barriers for the transfer of CSL into practice with threefold contributions: (1) We derive an information-theoretic conditional independence test that, incorporated into methods for CSL, improves the accuracy for non-linear and mixed discrete-continuous causal relationships; (2) We develop a modular pipeline that covers the essential components required for a comprehensive benchmarking to support the transferability into practice; (3) We evaluate opportunities and challenges of CSL within different real-world scenarios from genetics and discrete manufacturing to demonstrate the accuracy of our approach in practice.