Monthly Archives: June 2017

You are browsing the site archives by month.

Reinterpretation of evolutionary processes as algorithms for Bayesian inference

Jordan W. Suchow, David D. Bourgin, Thomas L. Griffiths, Evolution in Mind: Evolutionary Dynamics, Cognitive Processes, and Bayesian Inference, Trends in Cognitive Sciences, Volume 21, Issue 7, July 2017, Pages 522-530, ISSN 1364-6613, DOI: 10.1016/j.tics.2017.04.005.

Evolutionary theory describes the dynamics of population change in settings affected by reproduction, selection, mutation, and drift. In the context of human cognition, evolutionary theory is most often invoked to explain the origins of capacities such as language, metacognition, and spatial reasoning, framing them as functional adaptations to an ancestral environment. However, evolutionary theory is useful for understanding the mind in a second way: as a mathematical framework for describing evolving populations of thoughts, ideas, and memories within a single mind. In fact, deep correspondences exist between the mathematics of evolution and of learning, with perhaps the deepest being an equivalence between certain evolutionary dynamics and Bayesian inference. This equivalence permits reinterpretation of evolutionary processes as algorithms for Bayesian inference and has relevance for understanding diverse cognitive capacities, including memory and creativity.

A nice general model for camera calibration

S. Ramalingam and P. Sturm, “A Unifying Model for Camera Calibration,” in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 7, pp. 1309-1319, July 1 2017. DOI: 10.1109/TPAMI.2016.2592904.

This paper proposes a unified theory for calibrating a wide variety of camera models such as pinhole, fisheye, cata-dioptric, and multi-camera networks. We model any camera as a set of image pixels and their associated camera rays in space. Every pixel measures the light traveling along a (half-) ray in 3-space, associated with that pixel. By this definition, calibration simply refers to the computation of the mapping between pixels and the associated 3D rays. Such a mapping can be computed using images of calibration grids, which are objects with known 3D geometry, taken from unknown positions. This general camera model allows to represent non-central cameras; we also consider two special subclasses, namely central and axial cameras. In a central camera, all rays intersect in a single point, whereas the rays are completely arbitrary in a non-central one. Axial cameras are an intermediate case: the camera rays intersect a single line. In this work, we show the theory for calibrating central, axial and non-central models using calibration grids, which can be either three-dimensional or planar.

Clock synchronization in a wireless network based on consensus method that takes into account the noise (uncertainty) in the system, with a nice related work about consensus-based network clock synchronization

Jianping He, Xiaoming Duan, Peng Cheng, Ling Shi, Lin Cai, Accurate clock synchronization in wireless sensor networks with bounded noise, Automatica, Volume 81, July 2017, Pages 350-358, ISSN 0005-1098, DOI: 10.1016/j.automatica.2017.03.009.

It is important and challenging to achieve accurate clock synchronization in wireless sensor networks. Various noises, e.g., communication delay, clock fluctuation and measurement errors, are inevitable and difficult to be estimated accurately, which is the main challenge for achieving accurate clock synchronization. In this paper, we focus on how to achieve accurate clock synchronization by considering a practical noise model, bounded noise, which may not satisfy any known distributions. The principle that a bounded monotonic sequence must possess a limit and the concept of maximum consensus are exploited to design a novel clock synchronization algorithm for the network to achieve accurate and fast synchronization. The proposed algorithm is fully distributed, with high synchronization accuracy and fast convergence speed, and is able to compensate both clock skew and offset simultaneously. Meanwhile, we prove that the algorithm converges with probability one, which means that an accurate clock synchronization is achieved. We further prove that the probability of the complete synchronization converges exponentially fast. Experiments and simulations are conducted to verify the noise model and demonstrate the effectiveness of the proposed algorithm.

A good intro about actor-critic and decision making without model on MDPs

J. Wang and I. C. Paschalidis, “An Actor-Critic Algorithm With Second-Order Actor and Critic,” in IEEE Transactions on Automatic Control, vol. 62, no. 6, pp. 2689-2703, June 2017.DOI: 10.1109/TAC.2016.2616384.

Actor-critic algorithms solve dynamic decision making problems by optimizing a performance metric of interest over a user-specified parametric class of policies. They employ a combination of an actor, making policy improvement steps, and a critic, computing policy improvement directions. Many existing algorithms use a steepest ascent method to improve the policy, which is known to suffer from slow convergence for ill-conditioned problems. In this paper, we first develop an estimate of the (Hessian) matrix containing the second derivatives of the performance metric with respect to policy parameters. Using this estimate, we introduce a new second-order policy improvement method and couple it with a critic using a second-order learning method. We establish almost sure convergence of the new method to a neighborhood of a policy parameter stationary point. We compare the new algorithm with some existing algorithms in two applications and demonstrate that it leads to significantly faster convergence.

Interesting review of approaches to visually detect loop closings in robotics, and a novel, very efficient method that is independent on the image representation and based on not using the typical l2 norm (least squares), which leads to dense optimization problems

Yasir Latif, Guoquan Huang, John Leonard, José Neira, Sparse optimization for robust and efficient loop closing, Robotics and Autonomous Systems, Volume 93, July 2017, Pages 13-26, ISSN 0921-8890,DOI: 10.1016/j.robot.2017.03.016.

It is essential for a robot to be able to detect revisits or loop closures for long-term visual navigation. A key insight explored in this work is that the loop-closing event inherently occurs sparsely, i.e., the image currently being taken matches with only a small subset (if any) of previous images. Based on this observation, we formulate the problem of loop-closure detection as a sparse, convex
ℓ 1 -minimization problem. By leveraging fast convex optimization techniques, we are able to efficiently find loop closures, thus enabling real-time robot navigation. This novel formulation requires no offline dictionary learning, as required by most existing approaches, and thus allows online incremental operation. Our approach ensures a unique hypothesis by choosing only a single globally optimal match when making a loop-closure decision. Furthermore, the proposed formulation enjoys a flexible representation with no restriction imposed on how images should be represented, while requiring only that the representations are “close” to each other when the corresponding images are visually similar. The proposed algorithm is validated extensively using real-world datasets.

Evidences that the human brain has quantifying properties -i.e., ability to discriminate between sets of different sizes- as a result of evolution, but that numerical cognition is a result of culture

Rafael E. Núñez, Is There Really an Evolved Capacity for Number?, Trends in Cognitive Sciences, Volume 21, Issue 6, June 2017, Pages 409-424, ISSN 1364-6613, DOI: 10.1016/j.tics.2017.03.005.

Humans and other species have biologically endowed abilities for discriminating quantities. A widely accepted view sees such abilities as an evolved capacity specific for number and arithmetic. This view, however, is based on an implicit teleological rationale, builds on inaccurate conceptions of biological evolution, downplays human data from non-industrialized cultures, overinterprets results from trained animals, and is enabled by loose terminology that facilitates teleological argumentation. A distinction between quantical (e.g., quantity discrimination) and numerical (exact, symbolic) cognition is needed: quantical cognition provides biologically evolved preconditions for numerical cognition but it does not scale up to number and arithmetic, which require cultural mediation. The argument has implications for debates about the origins of other special capacities – geometry, music, art, and language.

Simultaneous localization and synchronization (SLAS) for multiple agents, with a nice state of the art including both SLAS for individual and multiple agents

B. Etzlinger, F. Meyer, F. Hlawatsch, A. Springer and H. Wymeersch, “Cooperative Simultaneous Localization and Synchronization in Mobile Agent Networks,” in IEEE Transactions on Signal Processing, vol. 65, no. 14, pp. 3587-3602, July15, 15 2017. DOI: 10.1109/TSP.2017.2691665.

Cooperative localization in agent networks based on interagent time-of-flight measurements is closely related to synchronization. To leverage this relation, we propose a Bayesian factor graph framework for cooperative simultaneous localization and synchronization (CoSLAS). This framework is suited to mobile agents and time-varying local clock parameters. Building on the CoSLAS factor graph, we develop a distributed (decentralized) belief propagation algorithm for CoSLAS in the practically important case of an affine clock model and asymmetric time stamping. Our algorithm is compatible with real-time operation and a time-varying network connectivity. To achieve high accuracy at reduced complexity and communication cost, the algorithm combines particle implementations with parametric message representations and takes advantage of a conditional independence property. Simulation results demonstrate the good performance of the proposed algorithm in a challenging scenario with time-varying network connectivity.

Modelling the implicit complexity of problem solving in exams

A. Shoufan, “Toward Modeling the Intrinsic Complexity of Test Problems,” in IEEE Transactions on Education, vol. 60, no. 2, pp. 157-163, May 2017.
DOI: 10.1109/TE.2016.2611666.

The concept of intrinsic complexity explains why different problems of the same type, tackled by the same problem solver, can require different times to solve and yield solutions of different quality. This paper proposes a general four-step approach that can be used to establish a model for the intrinsic complexity of a problem class in terms of solving time. Such a model allows prediction of the time to solve new problems in the same class and helps instructors develop more reliable test problems. A complexity model, furthermore, enhances understanding of the problem and can point to new aspects interesting for education and research. Students can use complexity models to assess and improve their learning level. The approach is explained using the K-map minimization problem as a case study. The implications of this research for other problems in electrical and computer engineering education are highlighted. An important aim of this paper is to stimulate future research in this area. An ideal outcome of such research is to provide complexity models for many, or even all, relevant problem classes in various electrical and computer engineering courses.

Personalizing the assessments generated automatically for students in order to minimize plagiarism: the case of programming

S. Manoharan, “Personalized Assessment as a Means to Mitigate Plagiarism,” in IEEE Transactions on Education, vol. 60, no. 2, pp. 112-119, May 2017.
DOI: 10.1109/TE.2016.2604210.

Although every educational institution has a code of academic honesty, they still encounter incidents of plagiarism. These are difficult and time-consuming to detect and deal with. This paper explores the use of personalized assessments with the goal of reducing incidents of plagiarism, proposing a personalized assessment software framework through which each student receives a unique problem set. The framework not only auto-generates the problem set but also auto-marks the solutions when submitted. The experience of using this framework is discussed, from the perspective of both students and staff, particularly with respect to its ability to mitigate plagiarism. A comparison of personalized and traditional assignments in the same class confirms that the former had far fewer observed plagiarism incidents. Although personalized assessment may not be cost-effective in all courses (such as language courses), it still can be effective in areas such as mathematics, engineering, science, and computing. This paper concludes that personalized assessment is a promising approach to counter plagiarism.

Reinforcement learning to learn the model of the world intrinsically motivated

Todd Hester, Peter Stone, Intrinsically motivated model learning for developing curious robots, Artificial Intelligence, Volume 247, June 2017, Pages 170-186, ISSN 0004-3702, DOI: 10.1016/j.artint.2015.05.002.

Reinforcement Learning (RL) agents are typically deployed to learn a specific, concrete task based on a pre-defined reward function. However, in some cases an agent may be able to gain experience in the domain prior to being given a task. In such cases, intrinsic motivation can be used to enable the agent to learn a useful model of the environment that is likely to help it learn its eventual tasks more efficiently. This paradigm fits robots particularly well, as they need to learn about their own dynamics and affordances which can be applied to many different tasks. This article presents the texplore with Variance-And-Novelty-Intrinsic-Rewards algorithm (texplore-vanir), an intrinsically motivated model-based RL algorithm. The algorithm learns models of the transition dynamics of a domain using random forests. It calculates two different intrinsic motivations from this model: one to explore where the model is uncertain, and one to acquire novel experiences that the model has not yet been trained on. This article presents experiments demonstrating that the combination of these two intrinsic rewards enables the algorithm to learn an accurate model of a domain with no external rewards and that the learned model can be used afterward to perform tasks in the domain. While learning the model, the agent explores the domain in a developing and curious way, progressively learning more complex skills. In addition, the experiments show that combining the agent’s intrinsic rewards with external task rewards enables the agent to learn faster than using external rewards alone. We also present results demonstrating the applicability of this approach to learning on robots.