Predicting earthquakes with precision remains an ongoing challenge in earthquake early warning systems (EEWS), that struggle with accuracy and fail to provide timely warnings for impending earthquakes. Recent efforts employing deep learning techniques have shown promise in overcoming these limitations. However, current methods lack the ability to capture subtle frequency changes indicative of seismic activity in real-time, limiting their effectiveness in EEWS. To address this gap, we propose REAVER, a novel approach for real-time prediction of P- and S-waves of earthquakes using attention-based sliding-window spectrograms. REAVER leverages Mel-Spectrogram signal representations to capture temporal frequency changes in seismic signals effectively. By employing an encoder-decoder architecture with attention mechanisms, REAVER accurately predicts the onset of P- and S-waves moments when an earthquake occurs. We benchmark the effectiveness of REAVER, showing its performance in terms of both accuracy and real-time prediction capabilities compared to existing methods. Additionally, we provide a web-based implementation of REAVER, allowing users to monitor seismic activity in real-time and analyze historical earthquake waveforms.
论文检索
输入标题、作者或关键词，从 1,048 篇学术成果中精准定位
Machine unlearning aims to remove specific knowledge from a well-trained machine learning model. This topic has gained significant attention recently due to the widespread adoption of machine learning models across various applications and the accompanying privacy, legal, and ethical considerations. During the unlearning process, models are typically presented with data that specifies which information should be erased and which should be retained. Nonetheless, practical challenges arise due to prevalent issues of data quality issues and access restrictions. This paper explores these challenges and introduces strategies to address problems related to unsupervised data, weakly supervised data, and scenarios characterized by zero-shot and federated data availability. Finally, we discuss related open questions, particularly concerning evaluation metrics, how the forgetting information is represented and delivered, and the unique challenges posed by large generative models.
This overview accompanies the author's Early Career Track presentation. We survey recent research and research agenda of the author, focusing on contributions in the area of computational argumentation. Contributions span from foundations of static and dynamic forms of argumentative reasoning and approaches to support explainability, e.g., analysis of the computational complexity of argumentative reasoning and algorithmic approaches.
In the area of symbolic AI, researchers strive to develop techniques to teach machines (commonsense) reasoning. Human reasoning is often argumentative in its nature, and consequently, computational models of argumentation constitute a vibrant research area in symbolic AI. In this paper I describe my most significant contributions to the field spanning from general non-monotonic logics to formal argumentation.
The field of social choice theory investigates how individual preferences are aggregated to reach collective decisions. While traditional social choice addresses problems such as choosing a winning candidate based on voter rankings or fairly allocating resources among individuals with the same entitlement, the wide range of decision-making scenarios in real-world applications calls for an extension beyond these basic frameworks. In this paper, I present an overview of my efforts to expand the reach of social choice theory in the domains of fair division, voting, and tournaments. Furthermore, I discuss avenues and challenges of bringing the developed theory closer to practice.
As data-driven AI models achieve unprecedented feats across previously unthinkable tasks, the diminishing levels of interpretability of their increasingly complex architectures can often be sidelined in place of performance. If we are to comprehend and trust these AI models as they advance, it is clear that symbolic methods, given their unparalleled strengths in knowledge representation and reasoning, can play an important role in explaining AI models. In this paper, I discuss some of the ways in which one branch of such methods, computational argumentation, given its human-like nature, can be used to tackle this problem. I first outline a general paradigm for this area of explainable AI, before detailing a prominent methodology therein which we have pioneered. I then illustrate how this approach has been put into practice with diverse AI models and types of explanations, before looking ahead to challenges, future work and the outlook in this field.
Recent advancements in recommender systems highlight the importance of metrics beyond accuracy, including diversity, serendipity, and fairness. This paper discusses various aspects of modern recommender systems, focusing on challenges such as preference elicitation, the complexity of human decision-making, and multi-domain applicability. The integration of Generative AI and Large Language Models offers enhanced personalization capabilities but also raises concerns regarding transparency and fairness. This work examines ongoing research efforts aimed at developing transparent, fair, and contextually aware systems. Our approach seeks to prioritize user wellbeing and responsibility, contributing to a more equitable and functional digital environment through advanced technologies and interdisciplinary insights.
Towards a Theory of Machine Learning on Graphs and its Applications in Combinatorial Optimization
PDF ↗Machine learning on graphs, especially using graph neural networks (GNNs), has seen a surge in interest due to the wide availability of graph data across many disciplines, from life and physical to social and engineering sciences. Despite their practical success, our theoretical understanding of the properties of GNNs remains incomplete. Here, we survey the author's and his collaborators' progress in developing a deeper theoretical understanding of GNNs' expressive power and generalization abilities. In addition, we overview recent progress in using GNNs to speed up solvers for hard combinatorial optimization tasks.
The Rise of Federated Intelligence: From Federated Foundation Models Toward Collective Intelligence
PDF ↗The success of foundation models advances the development of various intelligent and personalized agents to handle intricate tasks in their daily lives, however finite resources and privacy concerns from end users limit the potential of customizing the large intelligent agents for personal use. This paper explores the preliminary design of federated intelligence that paves the way toward personalized intelligent agents in large-scale collaboration scenarios. In Federated Intelligence, agents can collaboratively augment their intelligence quotient (IQ) by learning complementary knowledge and fine-grained adaptations. These personalized intelligent agents can also co-work together to jointly address complex tasks in the form of collective intelligence. The paper will highlight federated intelligence as a new pathway for tackling complex intelligent tasks by refining and extending centralized foundation models to an open and collaborative paradigm.
The widespread adoption of Artificial Intelligence (AI) systems has profoundly reshaped decision-making in social, political, and commercial contexts. This paper explores the critical issue of fairness in AI-driven decision-making, particularly in allocating resources and tasks. By examining recent advancements and key questions in computational social choice, I highlight challenges and prospects in designing fair systems in collective decision-making that are scalable, adaptable to intricate environments, and are aligned with complex and diverse human preferences.
Trustworthy machine learning (TML) under imperfect data has recently brought much attention in the data-centric fields of machine learning (ML) and artificial intelligence (AI). Specifically, there are mainly three types of imperfect data along with their challenges for ML, including i) label-level imperfection: noisy labels; ii) feature-level imperfection: adversarial examples; iii) distribution-level imperfection: out-of-distribution data. Therefore, in this paper, we systematically share our insights and solutions of TML to handle three types of imperfect data. More importantly, we discuss some new challenges in TML, which also open more opportunities for future studies, such as trustworthy foundation models, trustworthy federated learning, and trustworthy causal learning.
Human-AI Interaction Generation: A Connective Lens for Generative AI and Procedural Content Generation
PDF ↗Generative AI has recently gained popularity as a paradigm for content generation. In this paper, we link this paradigm to an older one: Procedural Content Generation (PCG). We propose a lens to identify the commonalities between both paradigms that we call human-AI interactive generation. Using this lens, we identify three beneficial attributes then survey recent related work and summarize relevant findings.
Human-Robot Alignment through Interactivity and Interpretability: Don't Assume a ``Spherical Human''
PDF ↗Interactive and interpretable robot learning can help to democratize robots, placing the power of assistive robotic systems in the hands of end-users. While machine learning-based approaches to robotics have achieved impressive results, robot learning is still a feat of costly engineering performed in controlled settings and relying upon impractical assumptions about humans. To achieve a vision in which robots can be integrated sustainably into our daily lives for robotic assistance, researchers must take a human-centered approach and develop novel approaches for human-robot alignment of robot values and behaviors. This paper amalgamates recent human factors insights and computational techniques that can support human-robot alignment through interactive and interpretable robot learning and teaming.
The primary research challenge lies in mitigating and measuring geographical and demographic biases in generative models, which is crucial for ensuring fairness in AI applications. Existing models trained on web-crawled datasets like LAION-400M often perpetuate harmful stereotypes and biases, especially concerning minority groups or less-represented regions. To address this, I proposed a framework called CIC (Culturally-aware Image Caption) to generate culturally-aware image captions. This framework leverages visual question answering (VQA) to extract cultural visual elements from images. It prompts both caption prompts and cultural visual elements to generate culturally-aware captions using large language models (LLMs). Human evaluations confirm the effectiveness of our approach in depicting cultural information accurately. Two key future directions are outlined. First, current image caption evaluation methods are inadequate for assessing culturally-aware captions, necessitating the development of new evaluation metrics leveraging cultural datasets and representations. Second, ethical considerations, particularly concerning stereotypes embedded in existing models, demand consensus and standards development through diverse cultural perspectives. Addressing these challenges is vital for the responsible deployment of AI technologies in diverse real-world contexts.
Current approaches to learning cooperative multi-agent behaviors assume relatively restrictive settings. In fully cooperative multi-agent reinforcement learning, the learning algorithm controls all agents in the scenario, while in ad hoc teamwork, the learning algorithm usually assumes control over only a single agent in the scenario. However, many cooperative settings in the real world are much less restrictive. For example, in an autonomous driving scenario, a company might train its cars to cooperate with each other, yet once on the road, these cars must additionally cooperate with cars from other companies. Towards expanding the class of scenarios that cooperative learning methods may optimally address, this research agenda introduces and proposes to study N-agent ad hoc teamwork (NAHT), where a set of autonomous agents must interact and cooperate with dynamically varying numbers and types of teammates.
In many allocation problems, understanding individual agents' needs, wants, and tradeoffs is crucial for providing fair and efficient solutions. This paper begins with motivating applications and critical definitions. We review existing results, such as advising agents on relaxing restrictions for improved resource allocation, optimizing task allocation in online settings without rejection of a task, and more. We conclude by outlining three potential directions for future research.
Auction-based federated learning (AFL) is an important area of FL incentive mechanism design. It effectively incentivizes high-quality data owners (DOs) to participate in data consumers' (DCs, i.e., servers') FL training tasks. However, AFL is still evolving, with existing methods primarily addressing optimal DC-DO matching or DC selection problems in monopoly markets. To enhance the practicality of AFL, we introduce stakeholder-oriented decision support in AFL. This facilitates optimal and strategic decision-making for all stakeholders, improving the efficiency and sustainability of the AFL ecosystem.
Bio-inspired Dynamic and Decentralized Online Learning in Uninformed Heterogeneous Multi-Agent Environments
PDF ↗Dynamic adaptation and learning, akin to natural organisms, is crucial for robots operating in real-world scenarios like search and rescue missions. We propose a solution combining intuition from embodied evolution and Bayes theory to promote flexible exploration in foraging tasks. Our investigation focuses on three main areas: 1) leveraging communication and prior knowledge to develop adaptable strategies in agent groups, 2) addressing challenges from sparse rewards or limited data availability, and 3) developing methods for concurrent evaluation and training in a single iteration, filling a current gap in learning-based solutions. Future directions include exploring decentralized coordination among agents and incorporating assistance based on prospective memory and altruism in multi-agent reinforcement learning.
Anomaly subgraph detection is a vital task in various real applications. However, with the advancement of AI technology, it faces new challenges: 1) Anomaly features are often deeply hidden within large datasets, and 2) Anomaly detection approaches are required to unveil the mechanisms behind anomaly generation. Our study focuses on detecting hidden anomaly subgraphs within big data and offering improved explanations for the root cause of anomalies by integrating multi-domain datasets.
Deep learning models have repeatedly shown their strengths in various application domains. However, their predictions often struggle to meet background knowledge requirements, which is a crucial condition for safety-critical systems. My research focuses on integrating requirements into neural networks to guide the learning process and ultimately produce outputs that ensure the requirements' satisfaction. Here, I will discuss my proposed methods in the context of two real-world applications: tabular data generation and autonomous driving.