AAAI 2023 Papers

Teaching young people about artificial intelligence (A.I.) is recognized globally as an important education effort by organizations and programs such as UNICEF, OECD, Elements of A.I., and AI4K12. A common theme among K-12 A.I. education programs is teaching how A.I. can impact society in both positive and negative ways. We present an effective tool that teaches young people about the societal impact of A.I. that goes one step further: empowering K-12 students to use tools and frameworks to create socially responsible A.I. The computational action process is a curriculum and toolkit that gives students the lessons and tools to evaluate positive and negative impacts of A.I. and consider how they can create beneficial solutions that involve A.I. and computing technology. In a human-subject research study, 101 U.S. and international students between ages 9 and 18 participated in a one-day workshop to learn and practice the computational action process. Pre-post questionnaires measured on the Likert scale students’ perception of A.I. in society and students' desire to use A.I. in their projects. Analysis of the results shows that students who identified as female agreed more strongly with having a concern about the impacts of A.I. than those who identified as male. Students also wrote open-ended responses to questions about what socially responsible technology means to them pre- and post-study. Analysis shows that post-intervention, students were more aware of ethical considerations and what tools they can use to code A.I. responsibly. In addition, students engaged actively with tools in the computational action toolkit, specifically the novel impact matrix, to describe the positive and negative impacts of A.I. technologies like facial recognition. Students demonstrated breadth and depth of discussion of various A.I. technologies' far-reaching positive and negative impacts. These promising results indicate that the computational action process can be a helpful addition to A.I. education programs in furnishing tools for students to analyze the effects of A.I. on society and plan how they can create and use socially responsible A.I.

We present nBIIG, a neural Business Intelligence (BI) Insights Generation system. Given a table, our system applies various analyses to create corresponding RDF representations, and then uses a neural model to generate fluent textual insights out of these representations. The generated insights can be used by an analyst, via a human-in-the-loop paradigm, to enhance the task of creating compelling table reports. The underlying generative neural model is trained over large and carefully distilled data, curated from multiple BI domains. Thus, the system can generate faithful and fluent insights over open-domain tables, making it practical and useful.

Human intelligence is multimodal; we integrate visual, linguistic, and acoustic signals to maintain a holistic worldview. Most current pretraining methods, however, are limited to one or two modalities. We present i-Code, a self-supervised pretraining framework where users may flexibly combine the modalities of vision, speech, and language into unified and general-purpose vector representations. In this framework, data from each modality are first given to pretrained single-modality encoders. The encoder outputs are then integrated with a multimodal fusion network, which uses novel merge- and co-attention mechanisms to effectively combine information from the different modalities. The entire system is pretrained end-to-end with new objectives including masked modality unit modeling and cross-modality contrastive learning. Unlike previous research using only video for pretraining, the i-Code framework can dynamically process single, dual, and triple-modality data during training and inference, flexibly projecting different combinations of modalities into a single representation space. Experimental results demonstrate how i-Code can outperform state-of-the-art techniques on five multimodal understanding tasks and single-modality benchmarks, improving by as much as 11% and demonstrating the power of integrative multimodal pretraining.

Low-resource relation extraction (LRE) aims to extract relations from limited labeled corpora. Existing work takes advantages of self-training or distant supervision to expand the limited labeled data in the data-driven approaches, while the selection bias of pseudo labels may cause the error accumulation in subsequent relation classification. To address this issue, this paper proposes fmLRE, an iterative feedback method based on feature mapping similarity calculation to improve the accuracy of pseudo labels. First, it calculates the similarities between pseudo-label and real-label data of the same category in a feature mapping space based on semantic features of labeled dataset after feature projection. Then, it fine-tunes initial model according to the iterative process of reinforcement learning. Finally, the similarity is used as a threshold for screening high-precision pseudo-labels and the basis for setting different rewards, which also acts as a penalty term for the loss function of relation classifier. Experimental results demonstrate that fmLRE achieves the state-of-the-art performance compared with strong baselines on two public datasets.

We present edBB-Demo, a demonstrator of an AI-powered research platform for student monitoring in remote education. The edBB platform aims to study the challenges associated to user recognition and behavior understanding in digital platforms. This platform has been developed for data collection, acquiring signals from a variety of sensors including keyboard, mouse, webcam, microphone, smartwatch, and an Electroencephalography band. The information captured from the sensors during the student sessions is modelled in a multimodal learning framework. The demonstrator includes: i) Biometric user authentication in an unsupervised environment; ii) Human action recognition based on remote video analysis; iii) Heart rate estimation from webcam video; and iv) Attention level estimation from facial expression analysis.

Electricity forecasting is crucial in scheduling and planning of future electric load, so as to improve the reliability and safeness of the power grid. Despite recent developments of forecasting algorithms in the machine learning community, there is a lack of general and advanced algorithms specifically considering requirements from the power industry perspective. In this paper, we present eForecaster, a unified AI platform including robust, flexible, and explainable machine learning algorithms for diversified electricity forecasting applications. Since Oct. 2021, multiple commercial bus load, system load, and renewable energy forecasting systems built upon eForecaster have been deployed in seven provinces of China. The deployed systems consistently reduce the average Mean Absolute Error (MAE) by 39.8% to 77.0%, with reduced manual work and explainable guidance. In particular, eForecaster also integrates multiple interpretation methods to uncover the working mechanism of the predictive models, which significantly improves forecasts adoption and user satisfaction.

Conventional temporal causal discovery (CD) methods suffer from high dimensionality, fail to identify lagged causal relationships, and often ignore dynamics in relations. In this study, we present a novel constraint-based CD approach for autocorrelated and non-stationary time series data (eCDANs) capable of detecting lagged and contemporaneous causal relationships along with temporal changes. eCDANs addresses high dimensionality by optimizing the conditioning sets while conducting conditional independence (CI) tests and identifies the changes in causal relations by introducing a surrogate variable to represent time dependency. Experiments on synthetic and real-world data show that eCDANs can identify time influence and outperform the baselines.

This paper addresses zero-shot slot filling, which tries to build a system that can generalize to unseen slot types without any training data. The key to zero-shot slot-filling is to match the tokens from the utterance with the semantic definition of the slot without training data in the target domain. This paper tackles this problem by devising a scheme to fully leverage pre-trained language models (PLMs). To this end, we propose a new prompting scheme that utilizes both learnable tokens and slot names to guide the model to focus on the relevant text spans for a given slot. Furthermore, we use attention values between tokens to form a feature descriptor for each token, which is motivated by the fact that the attention value in a PLM naturally characterizes various relationships, e.g., syntactic or semantic, between tokens. By further consolidating those features with an additional transformer-based aggregation module, we create a simple-but-effective zero-shot slot filling system that can achieve significantly better performance than the previous methods, as demonstrated by our experimental studies.

The spread of rumors along with breaking events seriously hinders the truth in the era of social media. Previous studies reveal that due to the lack of annotated resources, rumors presented in minority languages are hard to be detected. Furthermore, the unforeseen breaking events not involved in yesterday's news exacerbate the scarcity of data resources. In this work, we propose a novel zero-shot framework based on prompt learning to detect rumors falling in different domains or presented in different languages. More specifically, we firstly represent rumor circulated on social media as diverse propagation threads, then design a hierarchical prompt encoding mechanism to learn language-agnostic contextual representations for both prompts and rumor data. To further enhance domain adaptation, we model the domain-invariant structural features from the propagation threads, to incorporate structural position representations of influential community response. In addition, a new virtual response augmentation method is used to improve model training. Extensive experiments conducted on three real-world datasets demonstrate that our proposed model achieves much better performance than state-of-the-art methods and exhibits a superior capacity for detecting rumors at early stages.

Humans and animals engage in rich social interactions. It is often theorized that a relatively small number of basic social interactions give rise to the full range of behavior observed. But no computational theory explaining how social interactions combine together has been proposed before. We do so here. We take a model, the Social MDP, which is able to express a range of social interactions, and extend it to represent linear combinations of social interactions. Practically for robotics applications, such models are now able to not just express that an agent should help another agent, but to express goal-centric social interactions. Perhaps an agent is helping someone get dressed, but preventing them from falling, and is happy to exchange stories in the meantime. How an agent responds socially, should depend on what it thinks the other agent is doing at that point in time. To encode this notion, we take linear combinations of social interactions as defined in Social MDPs, and compute the weights on those combinations on the fly depending on the estimated goals of other agents. This new model, the Linear Social MDP, enables zero-shot reasoning about complex social interactions, provides a mathematical basis for the long-standing intuition that social interactions should compose, and leads to interesting new behaviors that we validate using human observers. Complex social interactions are part of the future of intelligent agents, and having principled mathematical models built on a foundation like MDPs will make it possible to bring social interactions to every robotic application.

Often a face has a voice. Appearance sometimes has a strong relationship with one's voice. In this work, we study how a face can be converted to a voice, which is a face-based voice conversion. Since there is no clean dataset that contains face and speech, voice conversion faces difficult learning and low-quality problems caused by background noise or echo. Too much redundant information for face-to-voice also causes synthesis of a general style of speech. Furthermore, previous work tried to disentangle speech with bottleneck adjustment. However, it is hard to decide on the size of the bottleneck. Therefore, we propose a bottleneck-free strategy for speech disentanglement. To avoid synthesizing the general style of speech, we utilize framewise facial embedding. It applied adversarial learning with a multi-scale discriminator for the model to achieve better quality. In addition, the self-attention module is added to focus on content-related features for in-the-wild data. Quantitative experiments show that our method outperforms previous work.

Event argument extraction (EAE) aims to identify the arguments of a given event, and classify the roles that those arguments play. Due to high data demands of training EAE models, zero-shot cross-lingual EAE has attracted increasing attention, as it greatly reduces human annotation effort. Some prior works indicate that generation-based methods have achieved promising performance for monolingual EAE. However, when applying existing generation-based methods to zero-shot cross-lingual EAE, we find two critical challenges, including Language Discrepancy and Template Construction. In this paper, we propose a novel method termed as Language-oriented Prefix-tuning Network (LAPIN) to address the above challenges. Specifically, we devise a Language-oriented Prefix Generator module to handle the discrepancies between source and target languages. Moreover, we leverage a Language-agnostic Template Constructor module to design templates that can be adapted to any language. Extensive experiments demonstrate that our proposed method achieves the best performance, outperforming the previous state-of-the-art model by 4.8% and 2.3% of the average F1-score on two multilingual EAE datasets.

We consider the problem of creating assistants that can help agents solve new sequential decision problems, assuming the agent is not able to specify the reward function explicitly to the assistant. Instead of acting in place of the agent as in current automation-based approaches, we give the assistant an advisory role and keep the agent in the loop as the main decision maker. The difficulty is that we must account for potential biases of the agent which may cause it to seemingly irrationally reject advice. To do this we introduce a novel formalization of assistance that models these biases, allowing the assistant to infer and adapt to them. We then introduce a new method for planning the assistant's actions which can scale to large decision making problems. We show experimentally that our approach adapts to these agent biases, and results in higher cumulative reward for the agent than automation-based alternatives. Lastly, we show that an approach combining advice and automation outperforms advice alone at the cost of losing some safety guarantees.

In classical planning, the aim is to find a sequence of deterministic actions leading from the initial to a goal state. In this work, we consider the scenario where a party who knows the solution to a planning task, called the prover, wants to convince a second party, the verifier, that it has the solution without revealing any information about the solution itself. This is relevant in domains where privacy is important, for example when plans contain sensitive information or when the solution should not be revealed upfront. We achieve this by introducing a zero-knowledge protocol for plan existence. By restricting ourselves to tasks with polynomially-bounded plan length, we are able to construct a protocol that can be run efficiently by both the prover and verifier. The resulting protocol does not rely on any reduction, has a constant number of rounds, and runs in time polynomial in the size of the task.

We formalize and analyze a fundamental component of dif- ferentiable neural architecture search (NAS): local “opera- tion scoring” at each operation choice. We view existing operation scoring functions as inexact proxies for accuracy, and we find that they perform poorly when analyzed empir- ically on NAS benchmarks. From this perspective, we intro- duce a novel perturbation-based zero-cost operation scor- ing (Zero-Cost-PT) approach, which utilizes zero-cost prox- ies that were recently studied in multi-trial NAS but de- grade significantly on larger search spaces, typical for dif- ferentiable NAS. We conduct a thorough empirical evalu- ation on a number of NAS benchmarks and large search spaces, from NAS-Bench-201, NAS-Bench-1Shot1, NAS- Bench-Macro, to DARTS-like and MobileNet-like spaces, showing significant improvements in both search time and accuracy. On the ImageNet classification task on the DARTS search space, our approach improved accuracy compared to the best current training-free methods (TE-NAS) while be- ing over 10× faster (total searching time 25 minutes on a single GPU), and observed significantly better transferabil- ity on architectures searched on the CIFAR-10 dataset with an accuracy increase of 1.8 pp. Our code is available at: https://github.com/zerocostptnas/zerocost operation score.

Traffic classification is a critical task in network security and management. Recent research has demonstrated the effectiveness of the deep learning-based traffic classification method. However, the following limitations remain: (1) the traffic representation is simply generated from raw packet bytes, resulting in the absence of important information; (2) the model structure of directly applying deep learning algorithms does not take traffic characteristics into account; and (3) scenario-specific classifier training usually requires a labor-intensive and time-consuming process to label data. In this paper, we introduce a masked autoencoder (MAE) based traffic transformer with multi-level flow representation to tackle these problems. To model raw traffic data, we design a formatted traffic representation matrix with hierarchical flow information. After that, we develop an efficient Traffic Transformer, in which packet-level and flow-level attention mechanisms implement more efficient feature extraction with lower complexity. At last, we utilize the MAE paradigm to pre-train our classifier with a large amount of unlabeled data, and perform fine-tuning with a few labeled data for a series of traffic classification tasks. Experiment findings reveal that our method outperforms state-of-the-art methods on five real-world traffic datasets by a large margin. The code is available at https://github.com/NSSL-SJTU/YaTC.

Video object detection (VID) is challenging because of the high variation of object appearance as well as the diverse deterioration in some frames. On the positive side, the detection in a certain frame of a video, compared with that in a still image, can draw support from other frames. Hence, how to aggregate features across different frames is pivotal to VID problem. Most of existing aggregation algorithms are customized for two-stage detectors. However, these detectors are usually computationally expensive due to their two-stage nature. This work proposes a simple yet effective strategy to address the above concerns, which costs marginal overheads with significant gains in accuracy. Concretely, different from traditional two-stage pipeline, we select important regions after the one-stage detection to avoid processing massive low-quality candidates. Besides, we evaluate the relationship between a target frame and reference frames to guide the aggregation. We conduct extensive experiments and ablation studies to verify the efficacy of our design, and reveal its superiority over other state-of-the-art VID approaches in both effectiveness and efficiency. Our YOLOX-based model can achieve promising performance (e.g., 87.5% AP50 at over 30 FPS on the ImageNet VID dataset on a single 2080Ti GPU), making it attractive for large-scale or real-time applications. The implementation is simple, we have made the demo codes and models available at https://github.com/YuHengsss/YOLOV.

Advances in artificial intelligence (AI) using techniques such as deep learning have fueled the recent progress in fields such as computer vision. However, these algorithms are still often viewed as "black boxes", which cannot easily explain how they arrived at their final output decisions. Saliency maps are one commonly used form of explainable AI (XAI), which indicate the input features an algorithm paid attention to during its decision process. Here, we introduce the open source xaitk-saliency package, an XAI framework and toolkit for saliency. We demonstrate its modular and flexible nature by highlighting two example use cases for saliency maps: (1) object detection model comparison and (2) doppelganger saliency for person re-identification. We also show how the xaitk-saliency package can be paired with visualization tools to support the interactive exploration of saliency maps. Our results suggest that saliency maps may play a critical role in the verification and validation of AI models, ensuring their trusted use and deployment. The code is publicly available at: https://github.com/xaitk/xaitk-saliency.

Recent development in the field of explainable artificial intelligence (XAI) has helped improve trust in Machine-Learning-as-a-Service (MLaaS) systems, in which an explanation is provided together with the model prediction in response to each query. However, XAI also opens a door for adversaries to gain insights into the black-box models in MLaaS, thereby making the models more vulnerable to several attacks. For example, feature-based explanations (e.g., SHAP) could expose the top important features that a black-box model focuses on. Such disclosure has been exploited to craft effective backdoor triggers against malware classifiers. To address this trade-off, we introduce a new concept of achieving local differential privacy (LDP) in the explanations, and from that we establish a defense, called XRand, against such attacks. We show that our mechanism restricts the information that the adversary can learn about the top important features, while maintaining the faithfulness of the explanations.