Association for the Advancement of Artificial Intelligence: AAAI Publications
Not a member yet
26155 research outputs found
Sort by
PersonalizedUS: Interpretable Breast Cancer Risk Assessment with Local Coverage Uncertainty Quantification
No full textCorrectly assessing the malignancy of breast lesions identified during ultrasound examinations is crucial for effective clinical decision-making. However, the current "gold standard" relies on manual BI-RADS scoring by clinicians, often leading to unnecessary biopsies and a significant mental health burden on patients and their families. In this paper, we introduce PersonalizedUS, an interpretable machine learning system that leverages recent advances in conformal prediction to provide precise and personalized risk estimates with local coverage guarantees and sensitivity, specificity, and predictive values above 0.9 across various threshold levels. In particular, we identify meaningful lesion subgroups where distribution-free, model-agnostic conditional coverage holds, with approximately 90% of our prediction sets containing only the ground truth in most lesion subgroups, thus explicitly characterizing for which patients the model is most suitably applied. Moreover, we make available a curated tabular dataset of 1936 biopsied breast lesions from a recent observational multicenter study and benchmark the performance of several state-of-the-art learning algorithms. We also report a successful case study of the deployed system in the same multicenter context. Concrete clinical benefits include up to a 65% reduction in requested biopsies among BI-RADS 4a and 4b lesions, with minimal to no missed cancer cases
ml4xcube: Machine Learning Toolkits for Earth System Data Cubes
No full textRapidly changing climate conditions and the increase in extreme events are posing severe challenges to human life and infrastructure, requiring sophisticated analytical capabilities for hazard prediction and disaster risk management. Earth System Data Cubes (ESDCs) have become an essential tool in Earth System Sciences (ESS) by organizing large-scale, multivariate environmental datasets into a structured, scalable and analysis-ready format. However, modern machine learning techniques are not yet being utilized to their full potential on ESDCs. This is due to the lack of proper tooling, domain-specific challenges, and high barriers of entry for practitioners. We introduce ml4xcube, an open-source Python framework designed to assist ESS domain experts in applying ML techniques on ESDCs for advanced analysis and prediction of environmental variables and impacts. Through a comprehensive suite of tools, it addresses specific challenges associated with the nature of ESS data, such as the non-uniform data distribution due to dynamic gaps, or spatio-temporal autocorrelation of environmental variables. Due to its modular architecture, it covers the complete analysis process, from data exploration, and preparation, to model development, result interpretation and evaluation. With support for distributed computing, it handles large ESDC datasets efficiently. In order to ease the adoption it includes extensive documentation and tutorial notebooks. We demonstrate ml4xcube's capabilities through three examples, showcasing its potential and capabilities for integrating machine learning with ESDC data
Multi-OphthaLingua: A Multilingual Benchmark for Assessing and Debiasing LLM Ophthalmological QA in LMICs
No full textCurrent ophthalmology clinical workflows are plagued by over-referrals, long waits, and complex and heterogeneous medical records. Large language models (LLMs) present a promising solution to automate various procedures such as triaging, preliminary tests like visual acuity assessment, and report summaries. However, LLMs have demonstrated significantly varied performance across different languages in natural language question-answering tasks, potentially exacerbating healthcare disparities in Low and Middle-Income Countries (LMICs). This study introduces the first multilingual ophthalmological question-answering benchmark with manually curated questions parallel across languages, allowing for direct cross-lingual comparisons. Our evaluation of 6 popular LLMs across 7 different languages reveals substantial bias across different languages, highlighting risks for clinical deployment of LLMs in LMICs. Existing debiasing methods such as Translation Chain-of-Thought or Retrieval-augmented generation (RAG) by themselves fall short of closing this performance gap, often failing to improve performance across all languages and lacking specificity for the medical domain. To address this issue, We propose CLARA (Cross-Lingual Reflective Agentic system), a novel inference time de-biasing method leveraging retrieval augmented generation and self-verification. Our approach not only improves performance across all languages but also significantly reduces the multilingual bias gap, facilitating equitable LLM application across the globe
MarkovType: A Markov Decision Process Strategy for Non-Invasive Brain-Computer Interfaces Typing Systems
No full textBrain-Computer Interfaces (BCIs) help people with severe speech and motor disabilities communicate and interact with their environment using neural activity. This work focuses on the Rapid Serial Visual Presentation (RSVP) paradigm of BCIs using noninvasive electroencephalography (EEG). The RSVP typing task is a recursive task with multiple sequences, where users see only a subset of symbols in each sequence. Extensive research has been conducted to improve classification in the RSVP typing task, achieving fast classification. However, these methods struggle to achieve high accuracy and do not consider the typing mechanism in the learning procedure. They apply binary target and non-target classification without including recursive training. To improve performance in the classification of symbols while controlling the classification speed, we incorporate the typing setup into training by proposing a Partially Observable Markov Decision Process (POMDP) approach. To the best of our knowledge, this is the first work to formulate the RSVP typing task as a POMDP for recursive classification. Experiments show that the proposed approach, MarkovType, results in a more accurate typing system compared to competitors. Additionally, our experiments demonstrate that while there is a trade-off between accuracy and speed, MarkovType achieves the optimal balance between these factors compared to other methods
GaraMoSt: Parallel Multi-Granularity Motion and Structural Modeling for Efficient Multi-Frame Interpolation in DSA Images
No full textThe rapid and accurate direct multi-frame interpolation method for Digital Subtraction Angiography (DSA) images is crucial for reducing radiation and providing real-time assistance to physicians for precise diagnostics and treatment. DSA images contain complex vascular structures and various motions. Applying natural scene Video Frame Interpolation (VFI) methods results in motion artifacts, structural dissipation, and blurriness. Recently, MoSt-DSA has specifically addressed these issues for the first time and achieved SOTA results. However, MoSt-DSA's focus on real-time performance leads to insufficient suppression of high-frequency noise and incomplete filtering of low-frequency noise in the generated images. To address these issues within the same computational time scale, we propose GaraMoSt. Specifically, we optimize the network pipeline with a parallel design and propose a module named MG-MSFE. MG-MSFE extracts frame-relative motion and structural features at various granularities in a fully convolutional parallel manner and supports independent, flexible adjustment of context-aware granularity at different scales, thus enhancing computational efficiency and accuracy. Extensive experiments demonstrate that GaraMoSt achieves the SOTA performance in accuracy, robustness, visual effects, and noise suppression, comprehensively surpassing MoSt-DSA and other natural scene VFI methods
Realizing AI for Impact: Towards Participatory Human-AI Collaboration for Water Conservation and Reproductive Health
No full textAI has immense potential for positive social impact, including in domains ranging from conservation to health. However, it can be challenging to account for human collaborations and real-world uncertainties when deploying such systems, which can lead to critical errors. Therefore, my research focuses on developing new methods in multi-agent systems and machine learning, including methods for participatory design of AI, human-AI collaboration, and uncertainty quantification, to develop safe, impactful AI systems, particularly in the domains of water conservation and reproductive health
Leveraging Human Input to Enable Robust, Interactive, and Aligned AI Systems
No full textEnsuring that AI systems do what we, as humans, actually want them to do, is one of the biggest open research challenges in AI alignment and safety. My research seeks to directly address this challenge by enabling AI systems to interact with humans to learn aligned and robust behaviors. The way in which robots and other AI systems behave is often the result of optimizing a reward function. However, manually designing good reward functions is highly challenging and error prone, even for domain experts. Consider trying to write down a reward function that describes good driving behavior or how you like your bed made in the morning. While reward functions for these tasks are difficult to manually specify, human feedback in the form of demonstrations or preferences are often much easier to obtain. However, human data is often difficult to interpret, due to ambiguity and noise. Thus, it is critical that AI systems take into account epistemic uncertainty over the human's true intent. My talk will give an overview of my lab's progress along the following fundamental research areas: (1) efficiently maintaining uncertainty over human intent, (2) directly optimizing behavior to be robust to uncertainty over human intent, and (3) actively querying for additional human input to reduce uncertainty over human intent
Learning Language Structures Through Grounding
No full textLanguage is highly structured, with syntactic and semantic structures, to some extent, agreed upon by speakers. With implicit or explicit awareness of such structures, humans can learn and use language efficiently and generalize to sentences that contain unseen words. Motivated by human language learning, in this presentation, I will introduce a family of machine learning tasks that learns language structures through grounding, where distant supervision from other data sources (i.e., grounds), including but not limited to different modalities (e.g., vision), execution results of programs, and other languages, are used to guide the learning of language structures. I will demonstrate the potential of this task formulation, advocate for its adoption through three schemes, and discuss the possibility of the general language learning problem through grounding
Compression-Aware Computing for Scalable and Sustainable AI
No full textThis talk explores the challenge of customizing large-scale AI models, particularly generative AI, on cost-effective devices with limited memory and energy resources. Modern AI models demand substantial computational power, often relying on specialized hardware such as GPUs. To address this, the talk introduces compression-aware computing, a framework enabling AI models to recognize and adapt to their compressed states while preserving performance. Compression-aware computing integrates compression techniques like sparsification, quantization, and low-rank decomposition to enhance the efficiency and accuracy of AI models, broadening these models' accessibility across diverse devices. Additionally, this talk highlights one rationale of scalable and sustainable AI in advancing Alzheimer’s research by facilitating the analysis of large single-cell transcriptomics datasets for gene-gene interaction discovery
Knowledge-driven Scientific Large Language Models
No full textMy research in AI for Science revolves around the development and application of knowledge graphs (KG) and large language models (LLM) for scientific discovery. Leveraging my expertise in AI, I extensively explore disciplinary knowledge, construct knowledge graphs, and develop pre-trained large models for chemical and biological research. The overarching goal is to better capture correlations and patterns between substances by incorporating explicit and implicit knowledge bases into pre-trained large models. I have published in top AI journals and conferences, including Nature Machine Intelligence, NeurIPS, AAAI, ICML, and ICLR, and received several prestigious awards such as the Excellent Prize of the Tencent Rhino-Bird Project (2024) and the Great Britain-China Educational Trust (2020). My research has garnered wide recognition, with over 6000 Google Scholar citations and GitHub repositories of my work on knowledge graph-enhanced molecular and protein learning receiving hundreds of stars. By pushing the boundaries of AI for scientific discovery, I aspire to contribute to significant advancements that address pressing global challenges. I am eager to present and share my work at AAAI’s New Faculty Highlight program and engage with fellow researchers at the forefront of AI