Search

Computational cognitive modeling is an important tool for understanding the processes supporting human and animal decision-making. Choice data in decision-making tasks are inherently noisy, and separating noise from signal can improve the quality of computational modeling. Common approaches to model decision noise often assume constant levels of noise or exploration throughout learning (e.g., the ϵ -softmax policy). However, this assumption is not guaranteed to hold – for example, a subject might disengage and lapse into an inattentive phase for a series of trials in the middle of otherwise low-noise performance. Here, we introduce a new, computationally inexpensive method to dynamically estimate the levels of noise fluctuations in choice behavior, under a model assumption that the agent can transition between two discrete latent states (e.g., fully engaged and random). Using simulations, we show that modeling noise levels dynamically instead of statically can substantially improve model fit and parameter estimation, especially in the presence of long periods of noisy behavior, such as prolonged lapses of attention. We further demonstrate the empirical benefits of dynamic noise estimation at the individual and group levels by validating it on four published datasets featuring diverse populations, tasks, and models. Based on the theoretical and empirical evaluation of the method reported in the current work, we expect that dynamic noise estimation will improve modeling in many decision-making paradigms over the static noise estimation method currently used in the modeling literature, while keeping additional model complexity and assumptions minimal. • Decision noise can fluctuate over time due to changes in task-engagement. • Cognitive models often fail to account for fluctuations in decision noise. • Our method models task-engagement as a binary, latent state. • We provide a procedure to estimate the latent state occupancy probability per trial. • Dynamic noise estimation can benefit model and parameter fit in theory and practice. [ABSTRACT FROM AUTHOR]

This paper considers a downlink distributed millimeter wave massive multi-input multi-output (D-mmMIMO) system, consisting of multiple randomly distributed radio access units (RAUs), the distribution of which is modeled as either a Poisson point process (PPP) or a Matérn hard-core point process (MHCPP). Compared with the PPP model where RAUs are randomly distributed without restriction, the MHCPP model constructed from a PPP by dependently thinning is more realistic, since it imposes a minimum distance between two RAUs so that they are not too close to each other. When a hybrid precoding algorithm based on antenna array response vectors is adopted, the upper bounds on average spectral efficiencies of the UEs for both the PPP D-mmMIMO system and MHCPP D-mmMIMO system are derived. Finally, simulation results are provided to demonstrate the validity of the developed analysis. The presented results also show that, when the MHCPP is generated from the PPP used in the PPP D-mmMIMO system, the MHCPP D-mmMIMO system with a low hard-core distance achieves a quite close performance to the PPP D-mmMIMO system, and increasing hard-core distance has a negative impact on spectral efficiency. In contrast, when the MHCPP is constructed from a different PPP, which system performs better depends on the intensities of two PPPs as well as hard core distance. [ABSTRACT FROM AUTHOR]

First-order Riesz transform based monogenic signal representation has been widely used in image processing and computer vision, however it only characterizes image intrinsic one-dimensional structure, and is incapable of describing intrinsic two-dimensional structure. To this end, a novel feature extraction approach, named Riesz Binary Pattern (RBP), is proposed for face recognition based on image multi-scale analysis and multi-order Riesz transform. RBP consists of two complementary components, i.e., local Riesz binary pattern (LRBP) and global Riesz binary pattern (GRBP). LRBP is obtained by performing local binary coding operator on each Riesz transform response to extract image intrinsic two-dimensional structure features. While GRBP is the global binary coding of joint information of image pixel multi-scale analysis and multi-order Riesz transform. Histogram of LRBP and GRBP are concatenated to form face image RBP description. Experimental results on three databases demonstrate that our proposed RBP descriptor is more discriminant in extracting image information and can provide a higher classification rate compared to some state-of-the-art image representation methods. [ABSTRACT FROM AUTHOR]

Sampling theorems for the two-dimensional linear canonical transform (LCT) in polar coordinates deserve special attention in medical imaging such as computerized tomography and magnetic resonance imaging due to the superiority of solving the traditional non-bandlimited images processing problems. Two types of interpolation formulae that interpolate in both radius and azimuth the angularly periodic and highest frequency limited functions with different bandwidth constraints, the LCT and the linear canonical Hankel transform (LCHT), are derived. The first interpolation formula is essentially equivalent to the latest one for bandlimited functions in the LCT domain while enjoys more concise and general mathematical derivations. Thanks to the consistency of the LCHT's order for bandlimited functions in the LCHT domain, the second interpolation formula outperforms the first one in computational complexity. [ABSTRACT FROM AUTHOR]