A Systems‐Level Approach to Address Risks and Ethics in Artificial Intelligence Systems
Systems Engineering, EarlyView.ABSTRACT Artificial intelligence (AI) is rapidly changing the world, from completely controlling routine or mundane tasks like text and image generation, to powering advanced algorithms that control critical systems. The recent advances in generative AI quickly overwhelmed multiple industries from education to finance as first adopters rushed (and ...
Vincent P. Paglioni, Torrey Mortenson
wiley +1 more source
HS-RankFormer for efficient and robust RGB-to-Hyperspectral image reconstruction across domains. [PDF]
Sci RepVamsi B +5 more
europepmc +1 more source
Wildlife Biology, EarlyView.Exploration of new wildlife surveying methodologies that leverage advances in sensor technology and machine learning has led to tentative research into the application of seismology techniques. This, most commonly, involves the deployment of a footfall trap – a seismic sensor and data logger customised for wildlife footfall.
Benjamin J. Blackledge +4 more
wiley +1 more source
StratGAN: Conditional Adversarial Network for Permittivity Inversion of Borehole Radar Data in Stratified Media. [PDF]
Sensors (Basel)Qing S +8 more
europepmc +1 more source
Reinforcement Learning With Timed Constraints for Robotics Motion Planning
Artificial Intelligence for Engineering, EarlyView.This work presents a unified automata‐based reinforcement learning framework that enforces MITL time‐bounded task specifications in both MDPs and POMDPs. Results from grid‐world and office scenarios show robust policy learning under stochastic dynamics and partial observability.
Zhaoan Wang +3 more
wiley +1 more source
Physics-Aware Spatiotemporal Consistency for Transferable Defense of Autonomous Driving Perception. [PDF]
Sensors (Basel)Liu Y +7 more
europepmc +1 more source
CAAI Transactions on Intelligence Technology, EarlyView.ABSTRACT As an attestation engagement, auditing is required to provide reasonable assurance for its conclusions. Traditional auditing has limited capacity to handle unstructured data and is usually based on audit sampling techniques, which can lead to the neglect of important audit evidence during the auditing process and result in a higher audit risk,
Xiaojia Wang, Ziqing Luo, Chaoxu Mu
wiley +1 more source
FP-ZOO: Fast Patch-Based Zeroth Order Optimization for Black-Box Adversarial Attacks on Vision Models. [PDF]
Sensors (Basel)Seo J, Jeon S.
europepmc +1 more source
Paired CycleGAN‐based virtual staining for 3D X‐ray histology of bone‐implant systems
Journal of Synchrotron Radiation, EarlyView.This work demonstrates virtual staining for 3D X‐ray histology using a modified CycleGAN trained on paired synchrotron radiation microtomography (SR‐µCT) and histology images of bone‐implant specimens. The method enables the generation of artificially coloured contrast resembling conventional histological staining from label‐free SR‐µCT data and is ...
Sarah C. Irvine +5 more
wiley +1 more source
Enhanced Magnetic Resonance Imaging-Based Knee Cartilage Segmentation Using a Swin-UNet Conditional Generative Adversarial Network: Development and Validation Study. [PDF]
JMIR Med InformPark JY +8 more
europepmc +1 more source