Bayesian Optimization Meets Laplace Approximation for Robotic Introspection
This addresses the issue of unreliable confidence estimates in deep learning for robotics, which is crucial for long-term autonomy, though it is an incremental improvement on existing methods.
The paper tackles the problem of deep neural networks lacking reliable confidence estimates in robotics by introducing a scalable Laplace Approximation technique to enable accurate failure probability assessments for unseen data. The result is a Bayesian Optimization algorithm that improves calibration and accuracy, requiring fewer iterations than random search and scaling to large datasets and architectures.
In robotics, deep learning (DL) methods are used more and more widely, but their general inability to provide reliable confidence estimates will ultimately lead to fragile and unreliable systems. This impedes the potential deployments of DL methods for long-term autonomy. Therefore, in this paper we introduce a scalable Laplace Approximation (LA) technique to make Deep Neural Networks (DNNs) more introspective, i.e. to enable them to provide accurate assessments of their failure probability for unseen test data. In particular, we propose a novel Bayesian Optimization (BO) algorithm to mitigate their tendency of under-fitting the true weight posterior, so that both the calibration and the accuracy of the predictions can be simultaneously optimized. We demonstrate empirically that the proposed BO approach requires fewer iterations for this when compared to random search, and we show that the proposed framework can be scaled up to large datasets and architectures.