Processing 是一门开源编程语言和与之配套的集成开发环境（IDE）的名称。Processing 在电子艺术和视觉设计社区被用来教授编程基础，并运用于大量的新媒体和互动艺术作品中。

Factor graphs have recently gained increasing attention as a unified framework for representing and constructing algorithms for signal processing, estimation, and control. One capability that does not seem to be well explored within the factor graph tool kit is the ability to handle deterministic nonlinear transformations, such as those occurring in nonlinear filtering and smoothing problems, using tabulated message passing rules. In this contribution, we provide general forward (filtering) and backward (smoothing) approximate Gaussian message passing rules for deterministic nonlinear transformation nodes in arbitrary factor graphs fulfilling a Markov property, based on numerical quadrature procedures for the forward pass and a Rauch-Tung-Striebel-type approximation of the backward pass. These message passing rules can be employed for deriving many algorithms for solving nonlinear problems using factor graphs, as is illustrated by the proposition of a nonlinear modified Bryson-Frazier (MBF) smoother based on the presented message passing rules.

In this paper we present mutual coinduction as a dual of mutual induction and also as a generalization of standard coinduction. In particular, we present a precise formal definition of mutual induction and mutual coinduction. In the process we present the associated mutual induction and mutual coinduction proof principles, and we present the conditions under which these principles hold. In spite of some mention of mutual (co)induction in research literature, but the formal definition of mutual (co)induction and the proof of the mutual (co)induction proof principles we present here seem to be the first such definition and proof. As such, to the best of our knowledge, it seems our work is the first to point out that, unlike the case for standard (co)induction, monotonicity of generators is not sufficient for guaranteeing the existence of least and greatest simultaneous fixed points in complete lattices, and that continuity on the other hand is sufficient for guaranteeing their existence. In the course of our presentation of mutual coinduction we also discuss some concepts related to standard (also called direct) induction and standard coinduction, as well as ones related to mutual (also called simultaneous or indirect) induction. During the presentation we purposely discuss particular standard concepts so as to help motivate the definitions of their more general counterparts for mutual/ simultaneous/ indirect (co)induction. Greatest simultaneous post-fixed points, in particular, will be abstractions and models of mathematical objects (e.g., points, sets, types, predicates, etc.) that are defined mutually-coinductively.

One of the largest obstacles facing scanning probe microscopy is the constant need to correct flaws in the scanning probe in situ. This is currently a manual, time-consuming process that would benefit greatly from automation. Here we introduce a convolutional neural network protocol that enables automated recognition of a variety of desirable and undesirable scanning probe tip states on both metal and non-metal surfaces. By combining the best performing models into majority voting ensembles, we find that the desirable states of H:Si(100) can be distinguished with a mean precision of 0.89 and an average receiver-operator-characteristic curve area of 0.95. More generally, high and low-quality tips can be distinguished with a mean precision of 0.96 and near perfect area-under-curve of 0.98. With trivial modifications, we also successfully automatically identify undesirable, non-surface-specific states on surfaces of Au(111) and Cu(111). In these cases we find mean precisions of 0.95 and 0.75 and area-under-curves of 0.98 and 0.94, respectively.

Thermal preferences vary from person to person and may change over time. The objective of this paper is to sequentially pose intelligent queries to occupants in order to optimally learn the room temperatures which maximize their satisfaction. Our central hypothesis is that an occupant's preference relation over room temperatures can be described using a scalar function of these temperatures, which we call the "occupant's thermal utility function". Information about an occupant's preference over room temperatures is available to us through their response to thermal preference queries : "prefer warmer," "prefer cooler" and "satisfied" which we interpret as statements about the derivative of their utility function, i.e. the utility function is "increasing", "decreasing" and "constant" respectively. We model this hidden utility function using a Gaussian process with a built-in unimodality constraint, i.e., the utility function has a unique maximum, and we train this model using Bayesian inference. This permits an expected improvement based selection of next preference query to pose to the occupant, which takes into account both exploration (sampling from areas of high uncertainty) and exploitation (sampling from areas which are likely to offer an improvement over current best observation). We use this framework to sequentially design experiments and illustrate its benefits by showing that it requires drastically fewer observations to learn the maximally preferred room temperature values as compared to other methods. This framework is an important step towards the development of intelligent HVAC systems which would be able to respond to individual occupants' personalized thermal comfort needs. In order to encourage the use of our PE framework and ensure reproducibility in results, we publish an implementation of our work named GPPrefElicit as an open-source package in the Python language .

In this paper we develop new confidence intervals for local polynomial regression, which minimize their worse-case coverage error and length in large samples. Our results rely on novel, valid Edgeworth expansions for $t$-statistics based on local polynomial methods, which are established uniformly over relevant classes of data generating processes and interval estimators. These higher-order expansions also allow for the uniform kernel and any derivative order, significantly improving on previous technical results available in the literature. In addition, we discuss principled, inference-optimal tuning parameter (bandwidth) selection and kernel functions. The main methodological results obtained in this paper are implemented in companion {\sf R} and \texttt{Stata} software packages.

A key feature of neural networks, particularly deep convolutional neural networks, is their ability to "learn" useful representations from data. The very last layer of a neural network is then simply a linear model trained on these "learned" representations. Despite their numerous applications in other tasks such as classification, retrieval, clustering etc., a.k.a. transfer learning, not much work has been published that investigates the structure of these representations or indeed whether structure can be imposed on them during the training process. In this paper, we study the effective dimensionality of the learned representations by models that have proved highly successful for image classification. We focus on ResNet-18, ResNet-50 and VGG-19 and observe that when trained on CIFAR10, CIFAR100 and SVHN, the learned representations exhibit a fairly low rank structure. We propose a modification to the training procedure, which further induces low rank structure on learned activations. Empirically, we show that this has implications for robustness to adversarial examples and compression.

The no-wait flowshop scheduling problem is a variant of the classical permutation flowshop problem, with the additional constraint that jobs have to be processed by the successive machines without waiting time. To efficiently address this NP-hard combinatorial optimization problem we conduct an analysis of the structure of good quality solutions. This analysis shows that the No-Wait specificity gives them a common structure: they share identical sub-sequences of jobs, we call super-jobs. After a discussion on the way to identify these super-jobs, we propose IG-SJ, an algorithm that exploits super-jobs within the state-of-the-art algorithm for the classical permutation flowshop, the well-known Iterated Greedy (IG) algorithm. An iterative approach of IG-SJ is also proposed. Experiments are conducted on Taillard's instances. The experimental results show that exploiting super-jobs is successful since IG-SJ is able to find 64 new best solutions.

Noisy images processing is a fundamental task of computer vision. The first example is the detection of faint edges in noisy images, a challenging problem studied in the last decades. A recent study introduced a fast method to detect faint edges in the highest accuracy among all the existing approaches. Their complexity is nearly linear in the image's pixels and their runtime is seconds for a noisy image. Their approach utilizes a multi-scale binary partitioning of the image. By utilizing the multi-scale U-net architecture, we show in this paper that their method can be dramatically improved in both aspects of run time and accuracy. By training the network on a dataset of binary images, we developed an approach for faint edge detection that works in a linear complexity. Our runtime of a noisy image is milliseconds on a GPU. Even though our method is orders of magnitude faster, we still achieve higher accuracy of detection under many challenging scenarios. In addition, we show that our approach to performing multi-scale preprocessing of noisy images using U-net improves the ability to perform other vision tasks under the presence of noise. We prove it on the problems of noisy objects classification and classical image denoising. We show that multi-scale denoising can be carried out by a novel edge preservation loss. As our experiments show, we achieve high-quality results in the three aspects of faint edge detection, noisy image classification and natural image denoising.

Neural network-based methods have recently demonstrated state-of-the-art results on image synthesis and super-resolution tasks, in particular by using variants of generative adversarial networks (GANs) with supervised feature losses. Nevertheless, previous feature loss formulations rely on the availability of large auxiliary classifier networks, and labeled datasets that enable such classifiers to be trained. Furthermore, there has been comparatively little work to explore the applicability of GAN-based methods to domains other than images and video. In this work we explore a GAN-based method for audio processing, and develop a convolutional neural network architecture to perform audio super-resolution. In addition to several new architectural building blocks for audio processing, a key component of our approach is the use of an autoencoder-based loss that enables training in the GAN framework, with feature losses derived from unlabeled data. We explore the impact of our architectural choices, and demonstrate significant improvements over previous works in terms of both objective and perceptual quality.

The present study provides a comparative assessment of non-invasive sensors as means of estimating the microbial contamination and time-on-shelf (i.e. storage time) of leafy green vegetables, using a novel unified spectra analysis workflow. Two fresh ready-to-eat green salads were used in the context of this study for the purpose of evaluating the efficiency and practical application of the presented workflow: rocket and baby spinach salads. The employed analysis workflow consisted of robust data normalization, powerful feature selection based on random forests regression, and selection of the number of partial least squares regression coefficients in the training process by estimating the knee-point on the explained variance plot. Training processes were based on microbiological and spectral data derived during storage of green salad samples at isothermal conditions (4, 8 and 12C), whereas testing was performed on data during storage under dynamic temperature conditions (simulating real-life temperature fluctuations in the food supply chain). Since an increasing interest in the use of non-invasive sensors in food quality assessment has been made evident in recent years, the unified spectra analysis workflow described herein, by being based on the creation/usage of limited sized featured sets, could be very useful in food-specific low-cost sensor development.

Signal machines form an abstract and idealised model of collision computing. Based on dimensionless signals moving on the real line, they model particle/signal dynamics in Cellular Automata. Each particle, or signal, moves at constant speed in continuous time and space. When signals meet, they get replaced by other signals. A signal machine defines the types of available signals, their speeds and the rules for replacement in collision. A signal machine A simulates another one B if all the space-time diagrams of B can be generated from space-time diagrams of A by removing some signals and renaming other signals according to local information. Given any finite set of speeds S, we construct a signal machine that is able to simulate any signal machine whose speeds belong to S. Each signal is simulated by a macro-signal, a ray of parallel signals. Each macro-signal has a main signal located exactly where the simulated signal would be, as well as auxiliary signals which encode its id and the collision rules of the simulated machine. The simulation of a collision, a macro-collision, consists of two phases. In the first phase, macro-signals are shrunk, then the macro-signals involved in the collision are identified and it is ensured that no other macro-signal comes too close. If some do, the process is aborted and the macro-signals are shrunk, so that the correct macro-collision will eventually be restarted and successfully initiated. Otherwise, the second phase starts: the appropriate collision rule is found and new macro-signals are generated accordingly. Considering all finite set of speeds S and their corresponding simulators provides an intrinsically universal family of signal machines.

This paper proposes using a linear function approximator, rather than a deep neural network (DNN), to bias a Monte Carlo tree search (MCTS) player for general games. This is unlikely to match the potential raw playing strength of DNNs, but has advantages in terms of generality, interpretability and resources (time and hardware) required for training. Features describing local patterns are used as inputs. The features are formulated in such a way that they are easily interpretable and applicable to a wide range of general games, and might encode simple local strategies. We gradually create new features during the same self-play training process used to learn feature weights. We evaluate the playing strength of an MCTS player biased by learnt features against a standard upper confidence bounds for trees (UCT) player in multiple different board games, and demonstrate significantly improved playing strength in the majority of them after a small number of self-play training games.

This work presents a dynamic power management architecture for neuromorphic many core systems such as SpiNNaker. A fast dynamic voltage and frequency scaling (DVFS) technique is presented which allows the processing elements (PE) to change their supply voltage and clock frequency individually and autonomously within less than 100 ns. This is employed by the neuromorphic simulation software flow, which defines the performance level (PL) of the PE based on the actual workload within each simulation cycle. A test chip in 28 nm SLP CMOS technology has been implemented. It includes 4 PEs which can be scaled from 0.7 V to 1.0 V with frequencies from 125 MHz to 500 MHz at three distinct PLs. By measurement of three neuromorphic benchmarks it is shown that the total PE power consumption can be reduced by 75%, with 80% baseline power reduction and a 50% reduction of energy per neuron and synapse computation, all while maintaining temporary peak system performance to achieve biological real-time operation of the system. A numerical model of this power management model is derived which allows DVFS architecture exploration for neuromorphics. The proposed technique is to be used for the second generation SpiNNaker neuromorphic many core system.

It is well-known that irreversible MCMC algorithms converge faster to their stationary distributions than reversible ones. Using the special geometric structure of Lie groups $\mathcal G$ and dissipation fields compatible with the symplectic structure, we construct an irreversible HMC-like MCMC algorithm on $\mathcal G$, where we first update the momentum by solving an OU process on the corresponding Lie algebra $\mathfrak g$, and then approximate the Hamiltonian system on $\mathcal G \times \mathfrak g$ with a reversible symplectic integrator followed by a Metropolis-Hastings correction step. In particular, when the OU process is simulated over sufficiently long times, we recover HMC as a special case. We illustrate this algorithm numerically using the example $\mathcal G = SO(3)$.

Top