Search

Predictions obtained by, e.g., artificial neural networks have a high accuracy but humans often perceive the models as black boxes. Insights about the decision making are mostly opaque for humans. Particularly understanding the decision making in highly sensitive areas such as healthcare or fifinance, is of paramount importance. The decision-making behind the black boxes requires it to be more transparent, accountable, and understandable for humans. This survey paper provides essential definitions, an overview of the different principles and methodologies of explainable Supervised Machine Learning (SML). We conduct a state-of-the-art survey that reviews past and recent explainable SML approaches and classifies them according to the introduced definitions. Finally, we illustrate principles by means of an explanatory case study and discuss important future directions., Comment: Accepted for publication at the Journal of Artificial Intelligence Research (JAIR)

We show that in the variational multiscale framework, the weak enforcement of essential boundary conditions via Nitsche's method corresponds directly to a particular choice of projection operator. The consistency, symmetry and penalty terms of Nitsche's method all originate from the fine-scale closure dictated by the corresponding scale decomposition. As a result of this formalism, we are able to determine the exact fine-scale contributions in Nitsche-type formulations. In the context of the advection-diffusion equation, we develop a residual-based model that incorporates the non-vanishing fine scales at the Dirichlet boundaries. This results in an additional boundary term with a new model parameter. We then propose a parameter estimation strategy for all parameters involved that is also consistent for higher-order basis functions. We illustrate with numerical experiments that our new augmented model mitigates the overly diffusive behavior that the classical residual-based fine-scale model exhibits in boundary layers at boundaries with weakly enforced essential conditions.

Artificial neural networks (NNs) have become the de facto standard in machine learning. They allow learning highly nonlinear transformations in a plethora of applications. However, NNs usually only provide point estimates without systematically quantifying corresponding uncertainties. In this paper a novel approach towards fully Bayesian NNs is proposed, where training and predictions of a perceptron are performed within the Bayesian inference framework in closed-form. The weights and the predictions of the perceptron are considered Gaussian random variables. Analytical expressions for predicting the perceptron's output and for learning the weights are provided for commonly used activation functions like sigmoid or ReLU. This approach requires no computationally expensive gradient calculations and further allows sequential learning., Comment: Accepted for publication at the 59th IEEE Conference on Decision and Control (CDC) 2020. v2: correction of typos

Recent years have seen many advances in methods for causal structure learning from data. The empirical assessment of such methods, however, is much less developed. Motivated by this gap, we pose the following question: how can one assess, in a given problem setting, the practical efficacy of one or more causal structure learning methods? We formalize the problem in a decision-theoretic framework, via a notion of expected loss or risk for the causal setting. We introduce a theoretical notion of causal risk as well as sample quantities that can be computed from data, and study the relationship between the two, both theoretically and through an extensive simulation study. Our results provide an assumptions-light framework for assessing causal structure learning methods that can be applied in a range of practical use-cases.

In this paper, we introduce a novel single shot approach for 6D object pose estimation of rigid objects based on depth images. For this purpose, a fully convolutional neural network is employed, where the 3D input data is spatially discretized and pose estimation is considered as a regression task that is solved locally on the resulting volume elements. With 65 fps on a GPU, our Object Pose Network (OP-Net) is extremely fast, is optimized end-to-end, and estimates the 6D pose of multiple objects in the image simultaneously. Our approach does not require manually 6D pose-annotated real-world datasets and transfers to the real world, although being entirely trained on synthetic data. The proposed method is evaluated on public benchmark datasets, where we can demonstrate that state-of-the-art methods are significantly outperformed., Comment: Accepted at 2020 IEEE International Conference on Robotics and Automation (ICRA 2020)

In this paper, we introduce a new public dataset for 6D object pose estimation and instance segmentation for industrial bin-picking. The dataset comprises both synthetic and real-world scenes. For both, point clouds, depth images, and annotations comprising the 6D pose (position and orientation), a visibility score, and a segmentation mask for each object are provided. Along with the raw data, a method for precisely annotating real-world scenes is proposed. To the best of our knowledge, this is the first public dataset for 6D object pose estimation and instance segmentation for bin-picking containing sufficiently annotated data for learning-based approaches. Furthermore, it is one of the largest public datasets for object pose estimation in general. The dataset is publicly available at http://www.bin-picking.ai/en/dataset.html., Comment: Accepted at 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2019)

For most industrial bin picking solutions, the pose of a workpiece is localized by matching a CAD model to point cloud obtained from 3D sensor. Distinguishing flat workpieces from bottom of the bin in point cloud imposes challenges in the localization of workpieces that lead to wrong or phantom detections. In this paper, we propose a framework that solves this problem by automatically segmenting workpiece regions from non-workpiece regions in a point cloud data. It is done in real time by applying a fully convolutional neural network trained on both simulated and real data. The real data has been labelled by our novel technique which automatically generates ground truth labels for real point clouds. Along with real time workpiece segmentation, our framework also helps in improving the number of detected workpieces and estimating the correct object poses. Moreover, it decreases the computation time by approximately 1s due to a reduction of the search space for the object pose estimation., Comment: IEEE CASE 2019