Neuromorphic event-based cameras are bio-inspired visual sensors with asynchronous pixels and extremely high temporal resolution. Such favorable properties make them an excellent choice for solving state estimation tasks under high-speed maneuvers. However, failures of camera pose tracking are frequently witnessed in state-of-the-art event-based visual odometry systems when the local map cannot be updated timely or feature matching is unreliable. One of the biggest roadblocks in this field is the absence of efficient and robust methods for data association without imposing any assumptions on the environment. This problem seems, however, unlikely to be addressed as in standard vision because of the motion-dependent nature of event data. To address this, we propose a map-free design for event-based visual-inertial state estimation in this paper. Instead of estimating camera position, we find that recovering the instantaneous linear velocity aligns better with event cameras' differential working principle. The proposed system uses raw data from a stereo event camera and an inertial measurement unit (IMU) as input, and adopts a dual-end architecture. The front-end preprocesses raw events and executes the computation of normal flow and depth information. To handle the temporally non-equispaced event data and establish association with temporally non-aligned IMU's measurements, the back-end employs a continuous-time formulation and a sliding-window scheme that can progressively estimate the linear velocity and IMU's bias. Experiments on synthetic and real data show our method achieves low-latency, metric-scale velocity estimation. To the best of our knowledge, this is the first real-time, purely event-based visual-inertial state estimator for high-speed maneuvers, requiring only sufficient textures and imposing no additional constraints on either the environment or motion pattern.

Event-based visual odometry is a specific branch of visual simultaneous localization and mapping (SLAM) techniques, which aims at solving tracking and mapping subproblems (typically in parallel), by exploiting the special working principles of neuromorphic (i.e., event-based) cameras. Due to the motion-dependent nature of event data, explicit data association (i.e., feature matching) under large-baseline viewpoint changes is difficult to establish, making direct methods a more rational choice. However, state-of-the-art direct methods are limited by the high computational complexity of the mapping subproblem and the degeneracy of camera pose tracking in certain degrees of freedom (DoF) in rotation. In this article, we tackle these issues by building an event-based stereo visual-inertial odometry system, which is built upon a direct pipeline known as event-based stereo visual odometry (ESVO). Specifically, to speed up the mapping operation, we propose an efficient strategy for sampling contour points according to the local dynamics of events. The mapping performance is also improved in terms of structure completeness and local smoothness by merging the temporal stereo and static stereo results. To circumvent the degeneracy of camera pose tracking in recovering the pitch and yaw components of general 6-DoF motion, we introduce IMU measurements as motion priors via preintegration. To this end, a compact back-end is proposed for continuously updating the IMU bias and predicting the linear velocity, enabling an accurate motion prediction for camera pose tracking. The resulting system scales well with modern high-resolution event cameras and leads to better global positioning accuracy in large-scale outdoor environments. Extensive evaluations on five publicly available datasets featuring different resolutions and scenarios justify the superior performance of the proposed system against five state-of-the-art methods. Compared to ESVO, our new pipeline significantly reduces the camera pose tracking error by 40%–80% and 20%–80% in terms of absolute trajectory error and relative pose error, respectively; at the same time, the mapping efficiency is improved by a factor of five. We release our pipeline as an open-source software for future research in this field.

Intelligence relies on an agent's knowledge of what it does not know. This capability can be assessed based on the quality of joint predictions of labels across multiple inputs. In principle, ensemble-based approaches can produce effective joint predictions, but the computational costs of large ensembles become prohibitive. We introduce the epinet: an architecture that can supplement any conventional neural network, including large pretrained models, and can be trained with modest incremental computation to estimate uncertainty. With an epinet, conventional neural networks outperform very large ensembles, consisting of hundreds or more particles, with orders of magnitude less computation. The epinet does not fit the traditional framework of Bayesian neural networks. To accommodate development of approaches beyond BNNs, such as the epinet, we introduce the epistemic neural network (ENN) as a general interface for models that produce joint predictions.

Thompson sampling (TS) is a popular heuristic for action selection, but it requires sampling from a posterior distribution. Unfortunately, this can become computationally intractable in complex environments, such as those modeled using neural networks. Approximate posterior samples can produce effective actions, but only if they reasonably approximate joint predictive distributions of outputs across inputs. Notably, accuracy of marginal predictive distributions does not suffice. Epistemic neural networks (ENNs) are designed to produce accurate joint predictive distributions. We compare a range of ENNs through computational experiments that assess their performance in approximating TS across bandit and reinforcement learning environments. The results indicate that ENNs serve this purpose well and illustrate how the quality of joint predictive distributions drives performance. Further, we demonstrate that the epinet – a small additive network that estimates uncertainty – matches the performance of large ensembles at orders of magnitude lower computational cost. This enables effective application of TS with computation that scales gracefully to complex environments.

Identifying independently moving objects is an essential task for dynamic scene understanding. However, traditional cameras used in dynamic scenes may suffer from motion blur or exposure artifacts due to their sampling principle. By contrast, event-based cameras are novel bio-inspired sensors that offer advantages to overcome such limitations. They report pixel-wise intensity changes asynchronously, which enables them to acquire visual information at exactly the same rate as the scene dynamics. We develop a method to identify independently moving objects acquired with an event-based camera, that is, to solve the event-based motion segmentation problem. We cast the problem as an energy minimization one involving the fitting of multiple motion models. We jointly solve two sub-problems, namely event-cluster assignment (labeling) and motion model fitting, in an iterative manner by exploiting the structure of the input event data in the form of a spatio-temporal graph. Experiments on available datasets demonstrate the versatility of the method in scenes with different motion patterns and number of moving objects. The evaluation shows state-of-the-art results without having to predetermine the number of expected moving objects. We release the software and dataset under an open source license to foster research in the emerging topic of event-based motion segmentation.

Predictive distributions quantify uncertainties ignored by point estimates. This paper introduces The Neural Testbed: an open source benchmark for controlled and principled evaluation of agents that generate such predictions. Crucially, the testbed assesses agents not only on the quality of their marginal predictions per input, but also on their joint predictions across many inputs. We evaluate a range of agents using a simple neural network data generating process.Our results indicate that some popular Bayesian deep learning agents do not fare well with joint predictions, even when they can produce accurate marginal predictions. We also show that the quality of joint predictions drives performance in downstream decision tasks. We find these results are robust across choice a wide range of generative models, and highlight the practical importance of joint predictions to the community.

Ensemble sampling serves as a practical approximation to Thompson sampling when maintaining an exact posterior distribution over model parameters is computationally intractable. In this paper, we establish a regret bound that ensures desirable behavior when ensemble sampling is applied to the linear bandit problem. This represents the first rigorous regret analysis of ensemble sampling and is made possible by leveraging information-theoretic concepts and novel analytic techniques that may prove useful beyond the scope of this paper.

Most work on supervised learning research has focused on marginal predictions. In decision problems, joint predictive distributions are essential for good performance. Previous work has developed methods for assessing low-order predictive distributions with inputs sampled i.i.d. from the testing distribution. With low-dimensional inputs, these methods distinguish agents that effectively estimate uncertainty from those that do not. We establish that the predictive distribution order required for such differentiation increases greatly with input dimension, rendering these methods impractical. To accommodate high-dimensional inputs, we introduce dyadic sampling, which focuses on predictive distributions associated with random pairs of inputs. We demonstrate that this approach efficiently distinguishes agents in high-dimensional examples involving simple logistic regression as well as complex synthetic and empirical data.

Visual–inertial odometry (VIO) is known to suffer from drifting, especially over long-term runs. In this article, we present GVINS, a nonlinear optimization-based system that tightly fuses global navigation satellite system (GNSS) raw measurements with visual and inertial information for real-time and drift-free stateestimation. Our system aims to provide accurate global six-degree-of-freedom estimation under complex indoor–outdoor environments, where GNSS signals may be intermittent or even inaccessible. To establish the connection between global measurements and local states, a coarse-to-fine initialization procedure is proposed to efficiently calibrate the transformation online and initialize GNSS states from only a short window of measurements. The GNSS code pseudorange and Doppler shift measurements, along with visual and inertial information, are then modeled and used to constrain the system states in a factor graph framework. For complex and GNSS-unfriendly areas, the degenerate cases are discussed and carefully handled to ensure robustness. Thanks to the tightly coupled multisensor approach and system design, our system fully exploits the merits of three types of sensors and is able to seamlessly cope with the transition between indoor and outdoor environments, where satellites are lost and reacquired. We extensively evaluate the proposed system by both simulation and real-world experiments, and the results demonstrate that our system substantially suppresses the drift of the VIO and preserves the local accuracy in spite of noisy GNSS measurements. The versatility and robustness of the system are verified on large-scale data collected in challenging environments. In addition, experiments show that our system can still benefit from the presence of only one satellite, whereas at least four satellites are required for its conventional GNSS counterparts.

Among prerequisites for a synthetic agent to inter-act with dynamic scenes, the ability to identify independently moving objects is specifically important. From an application perspective, nevertheless, standard cameras may deteriorate remarkably under aggressive motion and challenging illumination conditions. In contrast, event-based cameras, as a category of novel biologically inspired sensors, deliver advantages to deal with these challenges. Its rapid response and asynchronous nature enables it to capture visual stimuli at exactly the same rate of the scene dynamics. In this paper, we present a cascaded two-level multi-model fitting method for identifying independently moving objects (i.e., the motion segmentation problem) with a monocular event camera. The first level leverages tracking of event features and solves the feature clustering problem under a progressive multi-model fitting scheme. Initialized with the resulting motion model instances, the second level further addresses the event clustering problem using a spatio-temporal graph-cut method. This combination leads to efficient and accurate event-wise motion segmentation that cannot be achieved by any of them alone. Experiments demonstrate the effectiveness and versatility of our method in real-world scenes with different motion patterns and an unknown number of independently moving objects.

We integrate information-theoretic concepts into the design and analysis of optimistic algorithms and Thompson sampling. By making a connection between information-theoretic quantities and confidence bounds, we obtain results that relate the per-period performance of the agent with its information gain about the environment, thus explicitly characterizing the exploration-exploitation tradeoff. The resulting cumulative regret bound depends on the agent's uncertainty over the environment and quantifies the value of prior information. We show applicability of this approach to several environments, including linear bandits, tabular MDPs, and factored MDPs. These examples demonstrate the potential of a general information-theoretic approach for the design and analysis of reinforcement learning algorithms.

The low-rank structure is one of the most prominent features in modern recommendation problems. In this paper, we consider an online learning problem with a low-rank expected reward matrix where both row features and column features are unknown a priori, and the agent aims to learn to choose the best row-column pair (i.e. the maximum entry) in the matrix. We develop a novel online recommendation algorithm based on ensemble sampling, a recently developed computationally efficient approximation of Thompson sampling. Our computational results show that our algorithm consistently achieves order-of-magnitude improvements over the baselines in both synthetic and real-world experiments.

Thompson sampling has emerged as an effective heuristic for a broad range of online decision problems. In its basic form, the algorithm requires computing and sampling from a posterior distribution over models, which is tractable only for simple special cases. This paper develops ensemble sampling, which aims to approximate Thompson sampling while maintaining tractability even in the face of complex models such as neural networks. Ensemble sampling dramatically expands on the range of applications for which Thompson sampling is viable. We establish a theoretical basis that supports the approach and present computational results that offer further insight.