Search By Subject

Novel AI-generated audio samples are evaluated for descriptive qualities such as the smoothness of a morph using crowdsourced human listening tests. However, the methods to design interfaces for such experiments and to effectively articulate the ...

AI-driven Action Quality Assessment (AQA) of sports videos can mimic Olympic judges to help score performances as a second opinion or for training. However, these AI methods are uninterpretable and do not justify their scores, which is important for ...

As the state-of-the-art graph learning models, the message passing based neural networks (MPNNs) implicitly use the graph topology as the "pathways" to propagate node features. This implicit use of graph topology induces the MPNNs' over-reliance on (...

With the growing success of graph neural networks (GNNs), the explainability of GNN is attracting considerable attention. Current explainers mostly leverage feature attribution and selection to explain a prediction. By tracing the importance of input ...

Making each modality in multi-modal data contribute is of vital importance to learning a versatile multi-modal model. Existing methods, however, are often dominated by one or few of modalities during model training, resulting in sub-optimal performance. ...

Recent years witnessed several advances in developing multi-goal conversational recommender systems (MG-CRS) that can proactively attract users’ interests and naturally lead user-engaged dialogues with multiple conversational goals and diverse topics. ...

The availability of commercial wearable trackers equipped with features to monitor sleep duration and quality has enabled more useful sleep health monitoring applications and analyses. However, much research has reported the challenge of long-term user ...

Prediction of building energy consumption using machine learning models has been a focal point of research for decades. However, some causes of forecast errors, particularly data quality, have not been adequately addressed, which may affect the accuracy ...

Data-driven thermal comfort models play an important role in occupant-centric HVAC controls to satisfy occupants' preferences in thermal comfort. Although personal comfort models are likely to predict occupants' preferences well, the performance of ...

Audio-visual signals can be used jointly for robotic perception as they complement each other. Such multi-modal sensory fusion has a clear advantage, especially under noisy acoustic conditions. Speaker localization, as an essential robotic function, was ...

Dialogue state tracking (DST) is often used to track the system's understanding of the user goal in task-oriented dialogue systems. Existing DST methods mainly fall into two categories according to their adopted model structure: non-hierarchical ...

Classification tasks on labeled graph-structured data have many important applications ranging from social recommendation to financial modeling. Deep neural networks are increasingly being used for node classification on graphs, wherein nodes with ...

Membership inference (MI) attacks highlight a privacy weakness in present stochastic training methods for neural networks. It is not well understood, however, why they arise. Are they a natural consequence of imperfect generalization only? Which ...

How much does a machine learning algorithm leak about its training data, and why? Membership inference attacks are used as an auditing tool to quantify this leakage. In this paper, we present a comprehensivehypothesis testing framework that enables us ...

We introduce a new class of attacks on machine learning models. We show that an adversary who can poison a training dataset can cause models trained on this dataset to leak significant private details of training points belonging to other parties. Our ...

Despite the increasing popularity of VR games, one factor hindering the industry’s rapid growth is motion sickness experienced by the users. Symptoms such as fatigue and nausea severely hamper the user experience. Machine Learning methods could be used ...

Existing fact verification methods employ pre-trained language models such as BERT for the contextual representation of evidence sentences. However, such representations do not take into account commonsense knowledge and these methods often conclude ...

Financial texts (e.g., economic news) play an important role in predicting stock prices. The effects of texts of different semantics (e.g., launching a product and reporting a small product bug) last for different time horizons. Despite the importance ...

Pre-trained language model (LM) has led to significant performance gains in various natural language processing (NLP) applications due to its strong literacy, e.g., the ability to capture word dependencies. However, the existing pre-trained LMs largely ...

Given a stream of graph edges from a dynamic graph, how can we assign anomaly scores to edges in an online manner, for the purpose of detecting unusual behavior, using constant time and memory? Existing approaches aim to detect individually surprising ...