AAAI 2023 论文列表

This paper presents AnoViz, a novel visualization tool of anomalies in multivariate time series, to support domain experts and data scientists in understanding anomalous instances in their systems. AnoViz provides an overall summary of time series as well as detailed visualizations of relevant detected anomalies in both query and stream modes, rendering near real-time visual analysis available. Here, we show that AnoViz streamlines the process of finding a potential cause of an anomaly with a deeper analysis of anomalous instances, giving explainability to any anomaly detector.

In the light of the constant battle for attention on digital media, animating digital content plays an increasing role in modern graphic design. In this study, we use artificial intelligence methods to create aesthetic animations along the case of brand logos. With scalable vector graphics as the standard format in modern graphic design, we develop an autonomous end-to-end method using complex machine learning techniques to create brand logo animations as scalable vector graphics from scratch. We acquire data and setup a comprehensive animation space to create novel animations and evaluate them based on their aesthetics. We propose and compare two alternative computational models for automated logo animation and carefully weigh up their idiosyncrasies: on the one hand, we set up an aesthetics evaluation model to train an animation generator and, on the other hand, we combine tree ensembles with global optimization. Indeed, our proposed methods are capable of creating aesthetic logo animations, receiving an average rating of ‘good’ from observers.

The FastMap algorithm has been proposed as an inexpensive metric embedding which provides admissible distance estimates between all vertices in an embedding. As an embedding, it also supports additional operations such as taking the median location of two vertices, which is important in some problems. This paper studies several aspects of FastMap embeddings, showing the relationship of FastMap to general additive heuristics. As an admissible heuristic, FastMap is not as strong as previous suggested. However, by combining FastMap with the ideas of differential heuristics, we can significantly improve the performance of FastMap heuristics. We show the impact of these ideas in both single-agent pathfinding and the Multi-Agent Meeting problem, where the performance of algorithms using our improved FastMap embedding is improved by up to a factor of two.

Knowledge graph embedding (KGE), which maps entities and relations in a knowledge graph into continuous vector spaces, has achieved great success in predicting missing links in knowledge graphs. However, knowledge graphs often contain incomplete triples that are difficult to inductively infer by KGEs. To address this challenge, we resort to analogical inference and propose a novel and general self-supervised framework AnKGE to enhance KGE models with analogical inference capability. We propose an analogical object retriever that retrieves appropriate analogical objects from entity-level, relation-level, and triple-level. And in AnKGE, we train an analogy function for each level of analogical inference with the original element embedding from a well-trained KGE model as input, which outputs the analogical object embedding. In order to combine inductive inference capability from the original KGE model and analogical inference capability enhanced by AnKGE, we interpolate the analogy score with the base model score and introduce the adaptive weights in the score function for prediction. Through extensive experiments on FB15k-237 and WN18RR datasets, we show that AnKGE achieves competitive results on link prediction task and well performs analogical inference.

Analyzing the inner mechanisms of deep neural networks is a fundamental task in machine learning. Existing work provides limited analysis or it depends on local theories, such as fixed-point analysis. In contrast, we propose to analyze trained neural networks using an operator theoretic approach which is rooted in Koopman theory, the Koopman Analysis of Neural Networks (KANN). Key to our method is the Koopman operator, which is a linear object that globally represents the dominant behavior of the network dynamics. The linearity of the Koopman operator facilitates analysis via its eigenvectors and eigenvalues. Our method reveals that the latter eigendecomposition holds semantic information related to the neural network inner workings. For instance, the eigenvectors highlight positive and negative n-grams in the sentiments analysis task; similarly, the eigenvectors capture the salient features of healthy heart beat signals in the ECG classification problem.

In this study, we present a new presentation slide assessment system that can extract the structural features from any slide file formats. Our previous work used a neural network to identify novice vs. well-designed presentation slides based on visual and structural features. However, the structural feature extraction was only applicable to PowerPoint files. To solve this problem, we extract the semantic segmentation from the slide images as a new format of structural features. The proposed multi-modal Transformer extracts the features from the original images and semantic segmentation results to assess the slide design. The prediction targets are the top-10 checkpoints pointed out by the professional consultants. Class-imbalanced learning and multi-task learning methods are also applied to improve the accuracy. The proposed model only requiring the slide images achieved an average accuracy of 81.67% that is comparative to the performance of the previous work requiring the PowerPoint files.

The Feature Detection tool is a web-based activity that allows students to detect features in images and build their own rule-based classification algorithms. In this paper, we introduce the tool and share how it is incorporated into two, 45-minute lessons. The objective of the first lesson is to introduce students to the concept of feature detection, or how a computer can break down visual input into lower-level features. The second lesson aims to show students how these lower-level features can be incorporated into rule-based models to classify higher-order objects. We discuss how this tool can be used as a "first step" to the more complex concept ideas of data representation and neural networks.

Crowd-sourcing has attracted much attention due to its growing importance to society, and numerous studies have been conducted on task allocation and wage determination. Recent works have focused on optimizing task allocation and workers' wages, simultaneously. However, existing methods do not provide good solutions for real-world crowd-sourcing platforms due to the low approximation ratio or myopic problem settings. We tackle an optimization problem for wage determination and online task allocation in crowd-sourcing and propose a fast 1-1/(k+3)^(1/2)-approximation algorithm, where k is the minimum of tasks' budgets (numbers of possible assignments). This approximation ratio is greater than or equal to the existing method. The proposed method reduces the tackled problem to a non-convex multi-period continuous optimization problem by approximating the objective function. Then, the method transforms the reduced problem into a minimum convex cost flow problem, which is a well-known combinatorial optimization problem, and solves it by the capacity scaling algorithm. Synthetic experiments and simulation experiments using real crowd-sourcing data show that the proposed method solves the problem faster and outputs higher objective values than existing methods.

We study a fundamental model of online preference aggregation, where an algorithm maintains an ordered list of n elements. An input is a stream of preferred sets R_1, R_2, ..., R_t, ... Upon seeing R_t and without knowledge of any future sets, an algorithm has to rerank elements (change the list ordering), so that at least one element of R_t is found near the list front. The incurred cost is a sum of the list update costs (the number of swaps of neighboring list elements) and access cost (the position of the first element of R_t on the list). This scenario occurs naturally in applications such as ordering items in an online shop using aggregated preferences of shop customers. The theoretical underpinning of this problem is known as Min-Sum Set Cover. Unlike previous work that mostly studied the performance of an online algorithm ALG in comparison to the static optimal solution (a single optimal list ordering), in this paper, we study an arguably harder variant where the benchmark is the provably stronger optimal dynamic solution OPT (that may also modify the list ordering). In terms of an online shop, this means that the aggregated preferences of its user base evolve with time. We construct a computationally efficient randomized algorithm whose competitive ratio (ALG-to-OPT cost ratio) is O(r^2) and prove the existence of a deterministic O(r^4)-competitive algorithm. Here, r is the maximum cardinality of sets R_t. This is the first algorithm whose ratio does not depend on n: the previously best algorithm for this problem was O(r^(3/2) * n^(1/2))-competitive and Ω(r) is a lower bound on the performance of any deterministic online algorithm.

Forecasting time series with extreme events has been a challenging and prevalent research topic, especially when the time series data are affected by complicated uncertain factors, such as is the case in hydrologic prediction. Diverse traditional and deep learning models have been applied to discover the nonlinear relationships and recognize the complex patterns in these types of data. However, existing methods usually ignore the negative influence of imbalanced data, or severe events, on model training. Moreover, methods are usually evaluated on a small number of generally well-behaved time series, which does not show their ability to generalize. To tackle these issues, we propose a novel probability-enhanced neural network model, called NEC+, which concurrently learns extreme and normal prediction functions and a way to choose among them via selective back propagation. We evaluate the proposed model on the difficult 3-day ahead hourly water level prediction task applied to 9 reservoirs in California. Experimental results demonstrate that the proposed model significantly outperforms state-of-the-art baselines and exhibits superior generalization ability on data with diverse distributions.

We present a machine learning system for forecasting forced displacement populations deployed at the Danish Refugee Council (DRC). The system, named Foresight, supports long term forecasts aimed at humanitarian response planning. It is explainable, providing evidence and context supporting the forecast. Additionally, it supports scenarios, whereby analysts are able to generate forecasts under alternative conditions. The system has been in deployment since early 2020 and powers several downstream business functions within DRC. It is central to our annual Global Displacement Report which informs our response planning. We describe the system, key outcomes, lessons learnt, along with technical limitations and challenges in deploying machine learning systems in the humanitarian sector.

Quantification (or prevalence estimation) algorithms aim at predicting the class distribution of unseen sets (or bags) of examples. These methods are useful for two main tasks: 1) quantification applications, for instance when we need to track the proportions of several groups of interest over time, and 2) domain adaptation problems, in which we usually need to adapt a previously trained classifier to a different --albeit related-- target distribution according to the estimated prevalences. This paper analyzes several binary quantification algorithms showing that not only do they share a common framework but are, in fact, equivalent. Inspired by this study, we propose a new method that extends one of the approaches analyzed. After an empirical evaluation of all these methods using synthetic and benchmark datasets, the paper concludes recommending three of them due to their precision, efficiency, and diversity.

In this work, we propose a novel unsupervised contrastive learning framework to improve state-of-the-art sentence embeddings. First, we train a set of contrastive submodels which take multilingual round-trip translation(RTT) as data augmentation. The RTT naturally changes the length of the same sentence and replaces Synonyms simultaneously. Then we incorporate them into a single model through knowledge distillation. Specifically, it takes an input sentence and predicts the ensemble output of all submodels via a contrastive objective. Thus we preserve nearly the same semantic expressiveness as the ensemble model without increasing the test cost. We evaluate our framework on standard semantic textual similarity (STS) tasks. Experimental results show the advantage of our framework that we achieve an average of 79.27% Spearman's correlation, a 3.02% improvement compared to the previous best results using BERT-base.

Recent works on fake news detection have shown the efficacy of using emotions as a feature for improved performance. However, the cross-domain impact of emotion-guided features for fake news detection still remains an open problem. In this work, we propose an emotion-guided, domain-adaptive, multi-task approach for cross-domain fake news detection, proving the efficacy of emotion-guided models in cross-domain settings for various datasets.

One of the most popular methods for learning Nash equilibrium (NE) in large-scale imperfect information extensive-form games (IIEFGs) is the neural variants of counterfactual regret minimization (CFR). CFR is a special case of Follow-The-Regularized-Leader (FTRL). At each iteration, the neural variants of CFR update the agent's strategy via the estimated counterfactual regrets. Then, they use neural networks to approximate the new strategy, which incurs an approximation error. These approximation errors will accumulate since the counterfactual regrets at iteration t are estimated using the agent's past approximated strategies. Such accumulated approximation error causes poor performance. To address this accumulated approximation error, we propose a novel FTRL algorithm called FTRL-ORW, which does not utilize the agent's past strategies to pick the next iteration strategy. More importantly, FTRL-ORW can update its strategy via the trajectories sampled from the game, which is suitable to solve large-scale IIEFGs since sampling multiple actions for each information set is too expensive in such games. However, it remains unclear which algorithm to use to compute the next iteration strategy for FTRL-ORW when only such sampled trajectories are revealed at iteration t. To address this problem and scale FTRL-ORW to large-scale games, we provide a model-free method called Deep FTRL-ORW, which computes the next iteration strategy using model-free Maximum Entropy Deep Reinforcement Learning. Experimental results on two-player zero-sum IIEFGs show that Deep FTRL-ORW significantly outperforms existing model-free neural methods and OS-MCCFR.

Recently a multi-agent variant of the classical multi-armed bandit was proposed to tackle fairness issues in online learning. Inspired by a long line of work in social choice and economics, the goal is to optimize the Nash social welfare instead of the total utility. Unfortunately previous algorithms either are not efficient or achieve sub-optimal regret in terms of the number of rounds. We propose a new efficient algorithm with lower regret than even previous inefficient ones. We also complement our efficient algorithm with an inefficient approach with regret that matches the lower bound for one agent. The experimental findings confirm the effectiveness of our efficient algorithm compared to the previous approaches.

Active logic is a time-situated reasoner that can track the history of inferences, detect contradictions, and make parallel inferences in time. In this paper, we explore the behavior of an active-logic based agent on different sets of action selection axioms for a time-constrained target search task. We compare the performance of a baseline set of axioms that does not avoid redundant actions with five other axiom sets that avoid repeated actions but vary in their knowledge content. The results of these experiments show the importance of balancing boldness and caution for target search.

In light of significant issues in the technology industry, such as algorithms that worsen racial biases, the spread of online misinformation, and the expansion of mass surveillance, it is increasingly important to teach the ethics and sociotechnical implications of developing and using artificial intelligence (AI). Using 53 survey responses from engineering undergraduates, this paper measures students' abilities to identify, mitigate, and reflect on a hypothetical AI ethics scenario. We engage with prior research on pedagogical approaches to and considerations for teaching AI ethics and highlight some of the obstacles that engineering undergraduate students experience in learning and applying AI ethics concepts.

Many companies make use of customer service chats to help the customer and try to solve their problem. However, customer service data is confidential and as such, cannot easily be shared in the research community. This also implies that these data are rarely labeled, making it difficult to take advantage of it with machine learning methods. In this paper we present the first work on a customer’s problem status prediction and identification of problematic conversations. Given very small subsets of labeled textual conversations and unlabeled ones, we propose a semi-supervised framework dedicated to customer service data leveraging speaker role information to adapt the model to the domain and the task using a two-step process. Our framework, Task-Adaptive Fine-tuning, goes from predicting customer satisfaction to identifying the status of the customer’s problem, with the latter being the main objective of the multi-task setting. It outperforms recent inductive semi-supervised approaches on this novel task while only considering a relatively low number of parameters to train on during the final target task. We believe it can not only serve models dedicated to customer service but also to any other application making use of confidential conversational data where labeled sets are rare. Source code is available at https://github.com/gguibon/taft