Association for the Advancement of Artificial Intelligence: AAAI Publications
Not a member yet
26155 research outputs found
Sort by
Joint Scoring Rules: Competition Between Agents Avoids Performative Prediction
No full textIn a decision-making scenario, a principal could use conditional predictions from an expert agent to inform their choice. However, this approach would introduce a fundamental conflict of interest. An agent optimizing for predictive accuracy is incentivized to manipulate their principal towards more predictable actions, which prevents that principal from being able to deterministically select their true preference. We demonstrate that this impossibility result can be overcome through the joint evaluation of multiple agents. When agents are made to engage in zero-sum competition, their incentive to influence the action taken is eliminated, and the principal can identify and take the action they most prefer. We further prove that this zero-sum setup is unique, efficiently implementable, and applicable under stochastic choice. Experiments in a toy environment demonstrate that training on a zero-sum objective significantly enhances both predictive accuracy and principal utility, and can eliminate previously learned manipulative behavior
Text-Diffusion Red-Teaming of Large Language Models: Unveiling Harmful Behaviors with Proximity Constraints
No full textRecent work has proposed automated red-teaming methods for testing the vulnerabilities of a given target large language model (LLM). These methods use red-teaming LLMs to uncover inputs that induce harmful behavior in a target LLM.
In this paper, we study red-teaming strategies that enable a targeted security assessment. We propose an optimization framework for red-teaming with proximity constraints, where the discovered prompts must be similar to reference prompts from a given dataset. This dataset serves as a template for the discovered prompts, anchoring the search for test-cases to specific topics, writing styles, or types of harmful behavior. We show that established auto-regressive model architectures do not perform well in this setting.
We therefore introduce a black-box red-teaming method inspired by text-diffusion models: Diffusion for Auditing and Red-Teaming (DART). DART modifies the reference prompt by perturbing it in the embedding space, directly controlling the amount of change introduced. We systematically evaluate our method by comparing its effectiveness with established methods based on model fine-tuning and zero- and few-shot prompting. Our results show that DART is significantly more effective at discovering harmful inputs in close proximity to the reference prompt
To Measure or Not: A Cost-Sensitive, Selective Measuring Environment for Agricultural Management Decisions with Reinforcement Learning
No full textFarmers rely on in-field observations to make well-informed crop management decisions to maximize profit and minimize adverse environmental impact.
However, obtaining real-world crop state measurements is labor-intensive, time-consuming and expensive.
In most cases, it is not feasible to gather crop state measurements before every decision moment.
Moreover, in previous research pertaining to farm management optimization, these observations are often assumed to be readily available without any cost, which is unrealistic.
Hence, enabling optimization without the need to have *temporally complete* crop state observations is important.
An approach to that problem is to include measuring as part of decision making.
As a solution, we apply reinforcement learning (RL) to recommend opportune moments to simultaneously measure crop features and apply nitrogen fertilizer.
With realistic considerations, we design an RL environment with explicit crop feature measuring costs.
While balancing costs, we find that an RL agent, trained with recurrent PPO, discovers adaptive measuring policies that follow critical crop development stages, with results aligned by what domain experts would consider a sensible approach.
Our results highlight the importance of measuring when crop feature measurements are not readily available
Multi-Scale Graph Learning for Anti-Sparse Downscaling
No full textWater temperature can vary substantially even across short distances within the same sub-watershed. Accurate prediction of stream water temperature at fine spatial resolutions (i.e., fine scales, ≤ 1 km) enables precise interventions to maintain water quality and protect aquatic habitats. Although spatiotemporal models have made substantial progress in spatially coarse time series modeling, challenges persist in predicting at fine spatial scales due to the lack of data at that scale. To address the problem of insufficient fine-scale data, we propose a Multi-Scale Graph Learning (MSGL) method. This method employs a multi-task learning framework where coarse-scale graph learning, bolstered by larger datasets, simultaneously enhances fine-scale graph learning. Although existing multi-scale or multi-resolution methods integrate data from different spatial scales, they often overlook the spatial correspondences across graph structures at various scales. To address this, our MSGL introduces an additional learning task, cross-scale interpolation learning, which leverages the hydrological connectedness of stream locations across coarse- and fine-scale graphs to establish cross-scale connections, thereby enhancing overall model performance. Furthermore, we have broken free from the mindset that multi-scale learning is limited to synchronous training by proposing an Asynchronous Multi-Scale Graph Learning method (ASYNC-MSGL). Extensive experiments demonstrate the state-of-the-art performance of our method for anti-sparse downscaling of daily stream temperatures in the Delaware River Basin, USA, highlighting its potential utility for water resources monitoring and management
Constraint Optimisation Approaches for Designing Group-Living Captive Breeding Programmes
No full textCaptive breeding programs play a critical role in combating the ongoing biodiversity crisis by preserving the most endangered species and supporting reintroduction efforts. Maintaining the genetic health of captive populations requires careful management to prevent inbreeding and maximize the effective population size. Decisions about which males and females should be bred together are guided by the principle of minimizing relatedness between pairs. Methods to select breeding pairs are well developed, however, some species' ecology requires them to live in groups, and evaluating optimal groupings of multiple males and females that would be suitable to breed together is a more complex problem. Current computational tools to support the design of group-living captive breeding programs suffer from challenges of scalability and flexibility. In this paper we demonstrate the applicability of constraint programming (CP) approaches to optimize breeding groups to minimize relatedness. We present the example of the Galapagos giant tortoises as the test case used to develop our approach. Exploration of the needs of this captive breeding program has informed the development of our flexible approach to capture the constraints on viable captive breeding program design. Our findings have directly informed the implementation of new group configurations at the captive breeding centre. We further demonstrate that our approach is broadly applicable in other contexts through a second case study, providing multi-objective optimisation of a breeding program of canids. Through these case studies and an ablation study using synthetic datasets, we show that our constraint optimisation approach provides an expressive and generalizable means to support captive breeding program design, including scaling to large captive populations, which are currently intractable using current computational methods
Contextual Stochastic Optimization for School Desegregation Policymaking
No full textMost US school districts draw geographic "attendance zones" to
assign children to schools based on their home address, a process that
can replicate existing neighborhood racial/ethnic and socioeconomic
status (SES) segregation in schools. Redrawing boundaries can reduce
segregation, but estimating expected rezoning impacts is often challenging because
families can opt-out of their assigned schools. This paper seeks to
alleviate this societal problem by developing a joint
redistricting and choice modeling framework, called redistricting
with choices (RWC). The RWC framework is applied to a large US
public school district to estimate how redrawing elementary school
boundaries might realistically impact levels of
socioeconomic segregation. The main methodological contribution of
RWC is a contextual stochastic optimization model that aims to minimize
district-wide segregation by integrating rezoning constraints
with a machine learning-based school choice model. The study finds that RWC
yields boundary changes that might reduce segregation by a substantial
amount (23%) -- but doing so might require the re-assignment of a
large number of students, likely to mitigate re-segregation that
choice patterns could exacerbate. The results also reveal that
predicting school choice is a challenging machine learning problem.
Overall, this study offers a novel practical framework that both
academics and policymakers might use to foster more diverse and
integrated schools
Data Augmentation Approaches for Satellite Imagery
No full textDeep learning models commonly benefit from data augmentation techniques to diversify the set of training images. When working with satellite imagery, it is common for practitioners to apply a limited set of transformations developed for natural images (e.g., flip and rotate) to expand the training set without overly modifying the satellite images. There are many techniques for natural image data augmentation, but given the differences between the two domains, it is not clear whether data augmentation methods developed for natural images are well suited for satellite imagery. This paper presents an extensive experimental study on three classification and three regression tasks over four satellite image datasets. We compare common computer vision data augmentation techniques and propose three novel satellite-specific data augmentation strategies. Across tasks and datasets, we find that geometric transformations are beneficial for satellite imagery while color transformations generally are not. Additionally, our novel Sat-SlideMix, Sat-CutMix, and Sat-Trivial methods all exhibit strong performance across all tasks and datasets
Accessible, At-Home Detection of Parkinson’s Disease via Multi-Task Video Analysis
No full textLimited accessibility to neurological care leads to under-diagnosed Parkinson's Disease (PD), preventing early intervention. Existing AI-based PD detection methods primarily focus on unimodal analysis of motor or speech tasks, overlooking the multifaceted nature of the disease. To address this, we introduce a large-scale, multi-task video dataset of 1102 sessions (each containing videos of finger tapping, facial expression, and speech tasks captured via webcam) from 845 participants (272 with PD). We propose a novel Uncertainty-calibrated Fusion Network (UFNet) that leverages this multimodal data to enhance diagnostic accuracy. UFNet employs independent task-specific networks, trained with Monte Carlo Dropout for uncertainty quantification, followed by self-attended fusion of features, with attention weights dynamically adjusted based on task-specific uncertainties. We randomly split the participants into training (60%), validation (20%), and test (20%) sets to ensure patient-centered evaluation. UFNet significantly outperformed single-task models in terms of accuracy, area under the ROC curve (AUROC), and sensitivity while maintaining non-inferior specificity. Withholding uncertain predictions further boosted the performance, achieving 88.0 +- 0.3% accuracy, 93.0 +- 0.2% AUROC, 79.3 +- 0.9% sensitivity, and 92.6 +- 0.3% specificity, at the expense of not being able to predict for 2.3 +- 0.3% data (+- denotes 95% confidence interval). Further analysis suggests that the trained model does not exhibit any detectable bias across sex and ethnic subgroups and is most effective for individuals aged between 50 and 80. By merely requiring a webcam and microphone, our approach facilitates accessible home-based PD screening, especially in regions with limited healthcare resources
VidSole: A Multimodal Dataset for Joint Kinetics Quantification and Disease Detection with Deep Learning
No full textUnderstanding internal joint loading is critical for diagnosing gait-related diseases such as knee osteoarthritis; however, current methods of measuring joint risk factors are time-consuming, expensive, and restricted to lab settings. In this paper, we enable the large-scale, cost-effective biomechanical analysis of joint loading via three key contributions: the development and deployment of novel instrumented insoles, the creation of a large multimodal biomechanics dataset (VidSole), and a baseline deep learning pipeline to predict internal joint loading factors. Our novel instrumented insole measures the tri-axial forces and moments across five high-pressure points under the foot. VidSole consists of the forces and moments measured by these insoles along with corresponding RGB video from two viewpoints, 3D body motion capture, and force plate data for over 2,600 trials of 52 diverse participants performing four fundamental activities of daily living (sit-to-stand, stand-to-sit, walking, and running). We feed the insole data and kinematic parameters extractable from video (i.e., pose, knee angle) into a deep learning pipeline consisting of an ensemble Gated Recurrent Unit (GRU) activity classifier followed by activity-specific Long Short Term Memory (LSTM) regression networks to estimate knee adduction moment (KAM), a biomechanical risk factor for knee osteoarthritis. The successful classification of activities at an accuracy of 99.02 percent and KAM estimation with mean absolute error (MAE) less than 0.5 percent*body weight*height, the current threshold for accurately detecting knee osteoarthritis with KAM, illustrates the usefulness of our dataset for future research and clinical settings
Towards Efficient and Intelligent Laser Weeding: Method and Dataset for Weed Stem Detection
No full textWeed control is a critical challenge in modern agriculture, as weeds compete with crops for essential nutrient resources, significantly reducing crop yield and quality. Traditional weed control methods, including chemical and mechanical approaches, have real-life limitations such as associated environmental impact and efficiency.
An emerging yet effective approach is laser weeding, which uses a laser beam as the stem cutter. Although there have been studies that use deep learning in weed recognition, its application in intelligent laser weeding still requires a comprehensive understanding. Thus, this study serves the first empirical study on weed recognition for laser weeding. To increase the efficiency of laser beam cut and avoid damaging the crops of interest, the laser beam shall be directly aimed at the weed root. Yet, weed stem detection remains an under-explored problem. We integrate the detection of crop and weed with the localization of weed stem into one end-to-end system. To train and validate the proposed system in a real-life scenario, we curate and construct a high-quality weed stem detection dataset with human annotations. The dataset consists of 7,161 high-resolution pictures collected in the field with annotations of 11,151 instances of weed. The dataset will be released upon acceptance. Experimental results show that, in contrast to seminal weed recognition systems, the proposed system can efficiently improve the weeding accuracy by 5.05% and reduce the energy cost by 32.3%