Search

Cognitive scientists believe that adaptable intelligent agents like humans perform spatial reasoning tasks by learned causal mental simulation. The problem of learning these simulations is called predictive world modeling. We present the first framework for a learning open-vocabulary predictive world model (OV-PWM) from sensor observations. The model is implemented through a hierarchical variational autoencoder (HVAE) capable of predicting diverse and accurate fully observed environments from accumulated partial observations. We show that the OV-PWM can model high-dimensional embedding maps of latent compositional embeddings representing sets of overlapping semantics inferable by sufficient similarity inference. The OV-PWM simplifies the prior two-stage closed-set PWM approach to the single-stage end-to-end learning method. CARLA simulator experiments show that the OV-PWM can learn compact latent representations and generate diverse and accurate worlds with fine details like road markings, achieving 69 mIoU over six query semantics on an urban evaluation sequence. We propose the OV-PWM as a versatile continual learning paradigm for providing spatio-semantic memory and learned internal simulation capabilities to future general-purpose mobile robots. [ABSTRACT FROM AUTHOR]

The perception systems of autonomous vehicles face significant challenges under adverse conditions, with issues such as obscured objects and false detections due to environmental noise. Traditional approaches, which typically focus on noise removal, often fall short in such scenarios. Addressing the lack of diverse adverse weather data in existing automotive datasets, we propose a novel data augmentation method that integrates realistically simulated adverse weather effects into clear condition datasets. This method not only addresses the scarcity of data but also effectively bridges domain gaps between different driving environments. Our approach centers on a conditional generative model that uses segmentation maps as a guiding mechanism to ensure the authentic generation of adverse effects, which greatly enhances the robustness of perception and object detection systems in autonomous vehicles, operating under varied and challenging conditions. Besides the capability of accurately and naturally recreating over 90% of the adverse effects, we demonstrate that this model significantly improves the performance and accuracy of deep learning algorithms for autonomous driving, particularly in adverse weather scenarios. In the experiments employing our augmented approach, we achieved a 2.46% raise in the 3D average precision, a marked enhancement in detection accuracy and system reliability, substantiating the model's efficacy with quantifiable improvements in 3D object detection compared to models without augmentation. This work not only serves as an enhancement of autonomous vehicle perception systems under adverse conditions but also marked an advancement in deep learning models in adverse condition research. [ABSTRACT FROM AUTHOR]

Aim: Depressive disorder is often evaluated using established rating scales. However, consistent data collection with these scales requires trained professionals. In the present study, the "rater & estimation‐system" reliability was assessed between consensus evaluation by trained psychiatrists and the estimation by 2 models of the AI‐MADRS (Montgomery‐Asberg Depression Rating Scale) estimation system, a machine learning algorithm‐based model developed to assess the severity of depression. Methods: During interviews with trained psychiatrists and the AI‐MADRS estimation system, patients responded orally to machine‐generated voice prompts from the AI‐MADRS structured interview questions. The severity scores estimated from two models of the AI‐MADRS estimation system, the max estimation model and the average estimation model, were compared with those by trained psychiatrists. Results: A total of 51 evaluation interviews conducted on 30 patients were analyzed. Pearson's correlation coefficient with the scores evaluated by trained psychiatrists was 0.76 (95% confidence interval 0.62–0.86) for the max estimation model, and 0.86 (0.76–0.92) for the average estimation model. The ANOVA ICC rater & estimation‐system reliability with the evaluation scores by trained psychiatrists was 0.51 (−0.09 to 0.79) for the max estimation model, and 0.75 (0.55–0.86) for the average estimation model. Conclusion: The average estimation model of AI‐MADRS demonstrated substantially acceptable rater & estimation‐system reliability with trained psychiatrists. Accumulating a broader training dataset and the refinement of AI‐MADRS interviews are expected to improve the performance of AI‐MADRS. Our findings suggest that AI technologies can significantly modernize and potentially revolutionize the realm of depression assessments. [ABSTRACT FROM AUTHOR]