Controlling the Solo12 quadruped robot with deep reinforcement learning.

Quadruped robots require robust and general locomotion skills to exploit their mobility potential in complex and challenging environments. In this work, we present an implementation of a robust end-to-end learning-based controller on the Solo12 quadruped. Our method is based on deep reinforcement learning of joint impedance references. The resulting control policies follow a commanded velocity reference while being efficient in its energy consumption and easy to deploy. We detail the learning procedure and method for transfer on the real robot. We show elaborate experiments. Finally, we present experimental results of the learned locomotion on various grounds indoors and outdoors. These results show that the Solo12 robot is a suitable open-source platform for research combining learning and control because of the easiness in transferring and deploying learned controllers.

Unsupervised medical image classification by combining case-based classifiers.

We introduce an automated pathology classification system for medical volumetric brain image slices. Existing work often relies on handcrafted features extracted from automatic image segmentation. This is not only a challenging and time-consuming process, but it may also limit the adaptability and robustness of the system. We propose a novel approach to combine sparse Gabor-feature based classifiers in an ensemble classification framework. The unsupervised nature of this non-parametric technique can significantly reduce the time and effort for system calibration. In particular, classification of medical images in this framework does not rely on segmentation, nor semantic-based or annotation-based feature selection. Our experiments show very promising results in classifying computer tomography image slices into pathological classes for traumatic brain injury patients.

An automated pathological class level annotation system for volumetric brain images.

We introduce an automated, pathological class level annotation system for medical volumetric brain images. While much of the earlier work has mainly focused on annotating regions of interest in medical images, our system does not require annotated region level training data nor assumes perfect segmentation results for the regions of interest; the time and effort needed for acquiring training data are hence significantly reduced. This capability of handling high-dimensional noisy data, however, poses additional technical challenges, since statistical estimation of models for such data is prone to over-fitting. We propose a framework that combines a regularized logistic regression method and a kernel-based discriminative method to address these problems. Regularized methods provide a flexible selection mechanism that is well-suited for high dimensional noisy data. Our experiments show promising results in classifying computer tomography images of traumatic brain injury patients into pathological classes.

A generative model based approach to retrieving ischemic stroke images.

This paper proposes a generative model approach to automatically annotate medical images to improve the efficiency and effectiveness of image retrieval systems for teaching, research, and diagnosis. The generative model captures the probabilistic relationships among relevant classification tags, tentative lesion patterns, and selected input features. Operating on the imperfect segmentation results of input images, the probabilistic framework can effectively handle the inherent uncertainties in the images and insufficient information in the training data. Preliminary assessment in the ischemic stroke subtype classification shows that the proposed system is capable of generating the relevant tags for ischemic stroke brain images. The main benefit of this approach is its scalability; the method can be applied in large image databases as it requires only minimal manual labeling of the training data and does not demand high-precision segmentation of the images.

Association analyses of the MAS-QTL data set using grammar, principal components and Bayesian network methodologies.

BACKGROUND: It has been shown that if genetic relationships among individuals are not taken into account for genome wide association studies, this may lead to false positives. To address this problem, we used Genome-wide Rapid Association using Mixed Model and Regression and principal component stratification analyses. To account for linkage disequilibrium among the significant markers, principal components loadings obtained from top markers can be included as covariates. Estimation of Bayesian networks may also be useful to investigate linkage disequilibrium among SNPs and their relation with environmental variables.For the quantitative trait we first estimated residuals while taking polygenic effects into account. We then used a single SNP approach to detect the most significant SNPs based on the residuals and applied principal component regression to take linkage disequilibrium among these SNPs into account. For the categorical trait we used principal component stratification methodology to account for background effects. For correction of linkage disequilibrium we used principal component logit regression. Bayesian networks were estimated to investigate relationship among SNPs. RESULTS: Using the Genome-wide Rapid Association using Mixed Model and Regression and principal component stratification approach we detected around 100 significant SNPs for the quantitative trait (p<0.05 with 1000 permutations) and 109 significant (p<0.0006 with local FDR correction) SNPs for the categorical trait. With additional principal component regression we reduced the list to 16 and 50 SNPs for the quantitative and categorical trait, respectively. CONCLUSIONS: GRAMMAR could efficiently incorporate the information regarding random genetic effects. Principal component stratification should be cautiously used with stringent multiple hypothesis testing correction to correct for ancestral stratification and association analyses for binary traits when there are systematic genetic effects such as half sib family structures. Bayesian networks are useful to investigate relationships among SNPs and environmental variables.