This book from 1945 contains a very interesting mix of different charts made by the ISOTYPE Institute, some classic and some quite unusual. As a book about labor and unemployment, it also makes extensive use of Gerd Arntz’s famous unemployed man icon. Michael Young and Theodor Prager’s There’s Work for All is part of a […]

Visualization turns data into images, but are images themselves data? There are often claims that they are, but then you mostly see the images themselves without much additional data. In this video, I look at image browsers, a project classifying selfies along a number of criteria, and the additional information stored in HEIC that makes […]

Annotations are what set visual communication and journalism apart from just visualization. They often consist of text, but some of the most useful annotations are graphical elements, and many of them are very simple. One type I have a particular fondness for is the diagonal reference line, which has been used to provide powerful context […]

In this paper, we generalize the finiteness of models theorem in [BCHM06] to Kawamata log terminal pairs with fixed Kodaira dimension. As a consequence, we prove that a Kawamata log terminal pair with $mathbb{R}-$boundary has a canonical model, and can be approximated by log pairs with $mathbb{Q}-$boundary and the same canonical model.

For a smooth projective variety $X$, we study analogs of Quot functors in hearts of non-standard $t$-structures of $D^b(mathrm{Coh}(X))$. The technical framework is that of families of $t$-structures, as studied in arXiv:1902.08184. We provide several examples and suggest possible directions of further investigation, as we reinterpret moduli spaces of stable pairs, in the sense of Thaddeus (arXiv:alg-geom/9210007) and Huybrechts-Lehn (arXiv:alg-geom/9211001), as instances of Quot schemes.

We introduce the conormal fan of a matroid M, which is a Lagrangian analog of the Bergman fan of M. We use the conormal fan to give a Lagrangian interpretation of the Chern-Schwartz-MacPherson cycle of M. This allows us to express the h-vector of the broken circuit complex of M in terms of the intersection theory of the conormal fan of M. We also develop general tools for tropical Hodge theory to prove that the conormal fan satisfies Poincar'e duality, the hard Lefschetz theorem, and the Hodge-Riemann relations. The Lagrangian interpretation of the Chern-Schwartz-MacPherson cycle of M, when combined with the Hodge-Riemann relations for the conormal fan of M, implies Brylawski's and Dawson's conjectures that the h-vectors of the broken circuit complex and the independence complex of M are log-concave sequences.

Inserting renewable energy in the electric grid in a decentralized manneris a key challenge of the energy transition. However, at local scale, both production and demand display erratic behavior, which makes it delicate to match them. It is the goal of Energy Management Systems (EMS) to achieve such balance at least cost. We present EMSx, a numerical benchmark for testing control algorithms for the management of electric microgrids equipped with a photovoltaic unit and an energy storage system. EMSx is made of three key components: the EMSx dataset, provided by Schneider Electric, contains a diverse pool of realistic microgrids with a rich collection of historical observations and forecasts; the EMSx mathematical framework is an explicit description of the assessment of electric microgrid control techniques and algorithms; the EMSx software EMSx.jl is a package, implemented in the Julia language, which enables to easily implement a microgrid controller and to test it. All components of the benchmark are publicly available, so that other researchers willing to test controllers on EMSx may reproduce experiments easily. Eventually, we showcase the results of standard microgrid control methods, including Model Predictive Control, Open Loop Feedback Control and Stochastic Dynamic Programming.

We consider Landau-Ginzburg models stemming from groups comprised of non-diagonal symmetries, and we describe a rule for the mirror LG model. In particular, we present the non-abelian dual group, which serves as the appropriate choice of group for the mirror LG model. We also describe an explicit mirror map between the A-model and the B-model state spaces for two examples. Further, we prove that this mirror map is an isomorphism between the untwisted broad sectors and the narrow diagonal sectors for Fermat type polynomials.

We prove the effectiveness of the canonical bundle of several Hurwitz spaces of degree k covers of the projective line from curves of genus 13<g<20.

Let $f colon X o X$ be a surjective endomorphism of a normal projective surface. When $operatorname{deg} f geq 2$, applying an (iteration of) $f$-equivariant minimal model program (EMMP), we determine the geometric structure of $X$. Using this, we extend the second author's result to singular surfaces to the extent that either $X$ has an $f$-invariant non-constant rational function, or $f$ has infinitely many Zariski-dense forward orbits; this result is also extended to Adelic topology (which is finer than Zariski topology).

Let $X$ (resp. $Y$) be a curve of genus 1 (resp. 2) over a base field $k$ whose characteristic does not equal 2. We give criteria for the existence of a curve $Z$ over $k$ whose Jacobian is up to twist (2,2,2)-isogenous to the products of the Jacobians of $X$ and $Y$. Moreover, we give algorithms to construct the curve $Z$ once equations for $X$ and $Y$ are given. The first of these involves the use of hyperplane sections of the Kummer variety of $Y$ whose desingularization is isomorphic to $X$, whereas the second is based on interpolation methods involving numerical results over $mathbb{C}$ that are proved to be correct over general fields a posteriori. As an application, we find a twist of a Jacobian over $mathbb{Q}$ that admits a rational 70-torsion point.

We study the bi-commutators $[T_1, [b, T_2]]$ of pointwise multiplication and Calder'on-Zygmund operators, and characterize their $L^{p_1}L^{p_2} o L^{q_1}L^{q_2}$ boundedness for several off-diagonal regimes of the mixed-norm integrability exponents $(p_1,p_2) eq(q_1,q_2)$. The strategy is based on a bi-parameter version of the recent approximate weak factorization method.

Let $Gamma subset operatorname{PU}(1,n)$ be a lattice, and $S_Gamma$ the associated ball quotient. We prove that, if $S_Gamma$ contains infinitely many maximal totally geodesic subvarieties, then $Gamma$ is arithmetic. We also prove an Ax-Schanuel Conjecture for $S_Gamma$, similar to the one recently proven by Mok, Pila and Tsimerman. One of the main ingredients in the proofs is to realise $S_Gamma$ inside a period domain for polarised integral variations of Hodge structures and interpret totally geodesic subvarieties as unlikely intersections.

We construct a smooth moduli stack of tuples consisting of genus two nodal curves, line bundles, and twisted fields. It leads to a desingularization of the moduli of genus two stable maps to projective spaces. The construction of this new moduli is based on systematical application of the theory of stacks with twisted fields (STF), which has its prototype appeared in arXiv:1906.10527 and arXiv:1201.2427 and is fully developed in this article. The results of this article are the second step of a series of works toward the resolutions of the moduli of stable maps of higher genera.

In this paper we construct an equivariant embedding of the affine space $mathbb{A}^n$ with the translation group action into a complete non-projective algebraic variety $X$ for all $n geq 3$. The theory of toric varieties is used as the main tool for this construction. In the case of $n = 3$ we describe the orbit structure on the variety $X$.

We study a two-stage reneging queue with Poisson arrivals, exponential services, and two levels of exponential reneging behaviors, extending the popular Erlang A model that assumes a constant reneging rate. We derive approximate analytical formulas representing performance measures for the two-stage queue following the Markov chain decomposition approach. Our formulas not only give accurate results spanning the heavy-traffic to the light-traffic regimes, but also provide insight into capacity decisions.

We study SLE$_{kappa}$ theory with elements of Quasi-Sure Stochastic Analysis through Aggregation. Specifically, we show how the latter can be used to construct the SLE$_{kappa}$ traces quasi-surely (i.e. simultaneously for a family of probability measures with certain properties) for $kappa in mathcal{K}cap mathbb{R}_+ setminus ([0, epsilon) cup {8})$, for any $epsilon>0$ with $mathcal{K} subset mathbb{R}_{+}$ a nontrivial compact interval, i.e. for all $kappa$ that are not in a neighborhood of zero and are different from $8$. As a by-product of the analysis, we show in this language a version of the continuity in $kappa$ of the SLE$_{kappa}$ traces for all $kappa$ in compact intervals as above.

In this work we study the dynamical behavior Tonelli Lagrangian systems defined on the tangent bundle of the torus $mathbb{T}^2=mathbb{R}^2 / mathbb{Z}^2$. We prove that the Lagrangian flow restricted to a high energy level $ E_L^{-1}(c)$ (i.e $ c> c_0(L)$) has positive topological entropy if the flow satisfies the Kupka-Smale propriety in $ E_L^{-1}(c)$ (i.e, all closed orbit with energy $c$ are hyperbolic or elliptic and all heteroclinic intersections are transverse on $E_L^{-1}(c)$). The proof requires the use of well-known results in Aubry-Mather's Theory.

Input transformation based defense strategies fall short in defending against strong adversarial attacks. Some successful defenses adopt approaches that either increase the randomness within the applied transformations, or make the defense computationally intensive, making it substantially more challenging for the attacker. However, it limits the applicability of such defenses as a pre-processing step, similar to computationally heavy approaches that use retraining and network modifications to achieve robustness to perturbations. In this work, we propose a defense strategy that applies random image corruptions to the input image alone, constructs a self-correlation based subspace followed by a projection operation to suppress the adversarial perturbation. Due to its simplicity, the proposed defense is computationally efficient as compared to the state-of-the-art, and yet can withstand huge perturbations. Further, we develop proximity relationships between the projection operator of a clean image and of its adversarially perturbed version, via bounds relating geodesic distance on the Grassmannian to matrix Frobenius norms. We empirically show that our strategy is complementary to other weak defenses like JPEG compression and can be seamlessly integrated with them to create a stronger defense. We present extensive experiments on the ImageNet dataset across four different models namely InceptionV3, ResNet50, VGG16 and MobileNet models with perturbation magnitude set to {epsilon} = 16. Unlike state-of-the-art approaches, even without any retraining, the proposed strategy achieves an absolute improvement of ~ 4.5% in defense accuracy on ImageNet.

Large-scale pretrained language models are the major driving force behind recent improvements in performance on the Winograd Schema Challenge, a widely employed test of common sense reasoning ability. We show, however, with a new diagnostic dataset, that these models are sensitive to linguistic perturbations of the Winograd examples that minimally affect human understanding. Our results highlight interesting differences between humans and language models: language models are more sensitive to number or gender alternations and synonym replacements than humans, and humans are more stable and consistent in their predictions, maintain a much higher absolute performance, and perform better on non-associative instances than associative ones. Overall, humans are correct more often than out-of-the-box models, and the models are sometimes right for the wrong reasons. Finally, we show that fine-tuning on a large, task-specific dataset can offer a solution to these issues.

A standard approach to evaluating language models analyzes how models assign probabilities to valid versus invalid syntactic constructions (i.e. is a grammatical sentence more probable than an ungrammatical sentence). Our work uses ambiguous relative clause attachment to extend such evaluations to cases of multiple simultaneous valid interpretations, where stark grammaticality differences are absent. We compare model performance in English and Spanish to show that non-linguistic biases in RNN LMs advantageously overlap with syntactic structure in English but not Spanish. Thus, English models may appear to acquire human-like syntactic preferences, while models trained on Spanish fail to acquire comparable human-like preferences. We conclude by relating these results to broader concerns about the relationship between comprehension (i.e. typical language model use cases) and production (which generates the training data for language models), suggesting that necessary linguistic biases are not present in the training signal at all.

The connection between visual input and tactile sensing is critical for object manipulation tasks such as grasping and pushing. In this work, we introduce the challenging task of estimating a set of tactile physical properties from visual information. We aim to build a model that learns the complex mapping between visual information and tactile physical properties. We construct a first of its kind image-tactile dataset with over 400 multiview image sequences and the corresponding tactile properties. A total of fifteen tactile physical properties across categories including friction, compliance, adhesion, texture, and thermal conductance are measured and then estimated by our models. We develop a cross-modal framework comprised of an adversarial objective and a novel visuo-tactile joint classification loss. Additionally, we develop a neural architecture search framework capable of selecting optimal combinations of viewing angles for estimating a given physical property.

Device fingerprints like sensor pattern noise (SPN) are widely used for provenance analysis and image authentication. Over the past few years, the rapid advancement in digital photography has greatly reshaped the pipeline of image capturing process on consumer-level mobile devices. The flexibility of camera parameter settings and the emergence of multi-frame photography algorithms, especially high dynamic range (HDR) imaging, bring new challenges to device fingerprinting. The subsequent study on these topics requires a new purposefully built image dataset. In this paper, we present the Warwick Image Forensics Dataset, an image dataset of more than 58,600 images captured using 14 digital cameras with various exposure settings. Special attention to the exposure settings allows the images to be adopted by different multi-frame computational photography algorithms and for subsequent device fingerprinting. The dataset is released as an open-source, free for use for the digital forensic community.

Event-related desynchronization and synchronization (ERD/S) and movement-related cortical potential (MRCP) play an important role in brain-computer interfaces (BCI) for lower limb rehabilitation, particularly in standing and sitting. However, little is known about the differences in the cortical activation between standing and sitting, especially how the brain's intention modulates the pre-movement sensorimotor rhythm as they do for switching movements. In this study, we aim to investigate the decoding of continuous EEG rhythms during action observation (AO), motor imagery (MI), and motor execution (ME) for standing and sitting. We developed a behavioral task in which participants were instructed to perform both AO and MI/ME in regard to the actions of sit-to-stand and stand-to-sit. Our results demonstrated that the ERD was prominent during AO, whereas ERS was typical during MI at the alpha band across the sensorimotor area. A combination of the filter bank common spatial pattern (FBCSP) and support vector machine (SVM) for classification was used for both offline and pseudo-online analyses. The offline analysis indicated the classification of AO and MI providing the highest mean accuracy at 82.73$pm$2.38\% in stand-to-sit transition. By applying the pseudo-online analysis, we demonstrated the higher performance of decoding neural intentions from the MI paradigm in comparison to the ME paradigm. These observations led us to the promising aspect of using our developed tasks based on the integration of both AO and MI to build future exoskeleton-based rehabilitation systems.

Deep neural networks have exhibited promising performance in image super-resolution (SR). Most SR models follow a hierarchical architecture that contains both the cell-level design of computational blocks and the network-level design of the positions of upsampling blocks. However, designing SR models heavily relies on human expertise and is very labor-intensive. More critically, these SR models often contain a huge number of parameters and may not meet the requirements of computation resources in real-world applications. To address the above issues, we propose a Hierarchical Neural Architecture Search (HNAS) method to automatically design promising architectures with different requirements of computation cost. To this end, we design a hierarchical SR search space and propose a hierarchical controller for architecture search. Such a hierarchical controller is able to simultaneously find promising cell-level blocks and network-level positions of upsampling layers. Moreover, to design compact architectures with promising performance, we build a joint reward by considering both the performance and computation cost to guide the search process. Extensive experiments on five benchmark datasets demonstrate the superiority of our method over existing methods.

High-resolution remote sensing images (HRRSIs) contain substantial ground object information, such as texture, shape, and spatial location. Semantic segmentation, which is an important task for element extraction, has been widely used in processing mass HRRSIs. However, HRRSIs often exhibit large intraclass variance and small interclass variance due to the diversity and complexity of ground objects, thereby bringing great challenges to a semantic segmentation task. In this paper, we propose a new end-to-end semantic segmentation network, which integrates lightweight spatial and channel attention modules that can refine features adaptively. We compare our method with several classic methods on the ISPRS Vaihingen and Potsdam datasets. Experimental results show that our method can achieve better semantic segmentation results. The source codes are available at https://github.com/lehaifeng/SCAttNet.

For Convolutional Neural Network-based object detection, there is a typical dilemma: the spatial information is well kept in the shallow layers which unfortunately do not have enough semantic information, while the deep layers have a high semantic concept but lost a lot of spatial information, resulting in serious information imbalance. To acquire enough semantic information for shallow layers, Feature Pyramid Networks (FPN) is used to build a top-down propagated path. In this paper, except for top-down combining of information for shallow layers, we propose a novel network called Image Pyramid Guidance Network (IPG-Net) to make sure both the spatial information and semantic information are abundant for each layer. Our IPG-Net has two main parts: the image pyramid guidance transformation module and the image pyramid guidance fusion module. Our main idea is to introduce the image pyramid guidance into the backbone stream to solve the information imbalance problem, which alleviates the vanishment of the small object features. This IPG transformation module promises even in the deepest stage of the backbone, there is enough spatial information for bounding box regression and classification. Furthermore, we designed an effective fusion module to fuse the features from the image pyramid and features from the backbone stream. We have tried to apply this novel network to both one-stage and two-stage detection models, state of the art results are obtained on the most popular benchmark data sets, i.e. MS COCO and Pascal VOC.

Histopathology is a reflection of the molecular changes and provides prognostic phenotypes representing the disease progression. In this study, we introduced feature scores generated from hematoxylin and eosin histology images based on deep learning (DL) models developed for prostate pathology. We demonstrated that these feature scores were significantly prognostic for time to event endpoints (biochemical recurrence and cancer-specific survival) and had simultaneously molecular biologic associations to relevant genomic alterations and molecular subtypes using already trained DL models that were not previously exposed to the datasets of the current study. Further, we discussed the potential of such feature scores to improve the current tumor grading system and the challenges that are associated with tumor heterogeneity and the development of prognostic models from histology images. Our findings uncover the potential of feature scores from histology images as digital biomarkers in precision medicine and as an expanding utility for digital pathology.

In the era of Internet of Things (IoT), Malware has been proliferating exponentially over the past decade. Traditional anti-virus software are ineffective against modern complex Malware. In order to address this challenge, researchers have proposed Hardware-assisted Malware Detection (HMD) using Hardware Performance Counters (HPCs). The HPCs are used to train a set of Machine learning (ML) classifiers, which in turn, are used to distinguish benign programs from Malware. Recently, adversarial attacks have been designed by introducing perturbations in the HPC traces using an adversarial sample predictor to misclassify a program for specific HPCs. These attacks are designed with the basic assumption that the attacker is aware of the HPCs being used to detect Malware. Since modern processors consist of hundreds of HPCs, restricting to only a few of them for Malware detection aids the attacker. In this paper, we propose a Moving target defense (MTD) for this adversarial attack by designing multiple ML classifiers trained on different sets of HPCs. The MTD randomly selects a classifier; thus, confusing the attacker about the HPCs or the number of classifiers applied. We have developed an analytical model which proves that the probability of an attacker to guess the perfect HPC-classifier combination for MTD is extremely low (in the range of $10^{-1864}$ for a system with 20 HPCs). Our experimental results prove that the proposed defense is able to improve the classification accuracy of HPC traces that have been modified through an adversarial sample generator by up to 31.5%, for a near perfect (99.4%) restoration of the original accuracy.

In recent years there has been a burgeoning interest in the use of computational methods to distinguish between elicited speech samples produced by patients with dementia, and those from healthy controls. The difference between perplexity estimates from two neural language models (LMs) - one trained on transcripts of speech produced by healthy participants and the other trained on transcripts from patients with dementia - as a single feature for diagnostic classification of unseen transcripts has been shown to produce state-of-the-art performance. However, little is known about why this approach is effective, and on account of the lack of case/control matching in the most widely-used evaluation set of transcripts (DementiaBank), it is unclear if these approaches are truly diagnostic, or are sensitive to other variables. In this paper, we interrogate neural LMs trained on participants with and without dementia using synthetic narratives previously developed to simulate progressive semantic dementia by manipulating lexical frequency. We find that perplexity of neural LMs is strongly and differentially associated with lexical frequency, and that a mixture model resulting from interpolating control and dementia LMs improves upon the current state-of-the-art for models trained on transcript text exclusively.

Image dehazing is an ill-posed problem that has been extensively studied in the recent years. The objective performance evaluation of the dehazing methods is one of the major obstacles due to the lacking of a reference dataset. While the synthetic datasets have shown important limitations, the few realistic datasets introduced recently assume homogeneous haze over the entire scene. Since in many real cases haze is not uniformly distributed we introduce NH-HAZE, a non-homogeneous realistic dataset with pairs of real hazy and corresponding haze-free images. This is the first non-homogeneous image dehazing dataset and contains 55 outdoor scenes. The non-homogeneous haze has been introduced in the scene using a professional haze generator that imitates the real conditions of hazy scenes. Additionally, this work presents an objective assessment of several state-of-the-art single image dehazing methods that were evaluated using NH-HAZE dataset.

Recent studies have confirmed a growing trend, especially among youngsters, of using Online Social Media as favourite information platform at the expense of traditional mass media. Indeed, they can easily reach a wide audience at a high speed; but exactly because of this they are the preferred medium for influencing public opinion via so-called fake news. Moreover, there is a general agreement that the main vehicle of fakes news are malicious software robots (bots) that automatically interact with human users. In previous work we have considered the problem of tagging human users in Online Social Networks as credulous users. Specifically, we have considered credulous those users with relatively high number of bot friends when compared to total number of their social friends. We consider this group of users worth of attention because they might have a higher exposure to malicious activities and they may contribute to the spreading of fake information by sharing dubious content. In this work, starting from a dataset of fake news, we investigate the behaviour and the degree of involvement of credulous users in fake news diffusion. The study aims to: (i) fight fake news by considering the content diffused by credulous users; (ii) highlight the relationship between credulous users and fake news spreading; (iii) target fake news detection by focusing on the analysis of specific accounts more exposed to malicious activities of bots. Our first results demonstrate a strong involvement of credulous users in fake news diffusion. This findings are calling for tools that, by performing data streaming on credulous' users actions, enables us to perform targeted fact-checking.

Participatory Design's vision of democratic participation assumes participants' feelings of agency in envisioning a collective future. But this assumption may be leaky when dealing with vulnerable populations. We reflect on the results of a series of activities aimed at supporting agentic-future-envisionment with a group of sex-trafficking survivors in Nepal. We observed a growing sense among the survivors that they could play a role in bringing about change in their families. They also became aware of how they could interact with available institutional resources. Reflecting on the observations, we argue that building participant agency on the small and personal interactions is necessary before demanding larger Political participation. In particular, a value of PD, especially for vulnerable populations, can lie in the process itself if it helps participants position themselves as actors in the larger world.

One of the main open problems in the qualitative theory of real planar differential systems is the study of limit cycles. In this article, we present an algorithmic approach for detecting how many limit cycles can bifurcate from the periodic orbits of a given polynomial differential center when it is perturbed inside a class of polynomial differential systems via the averaging method. We propose four symbolic algorithms to implement the averaging method. The first algorithm is based on the change of polar coordinates that allows one to transform a considered differential system to the normal form of averaging. The second algorithm is used to derive the solutions of certain differential systems associated to the unperturbed term of the normal of averaging. The third algorithm exploits the partial Bell polynomials and allows one to compute the integral formula of the averaged functions at any order. The last algorithm is based on the aforementioned algorithms and determines the exact expressions of the averaged functions for the considered differential systems. The implementation of our algorithms is discussed and evaluated using several examples. The experimental results have extended the existing relevant results for certain classes of differential systems.

Convolution is an equivariant operation, and image position does not affect its result. A recent study shows that the zero-padding employed in convolutional layers of CNNs provides position information to the CNNs. The study further claims that the position information enables accurate inference for several tasks, such as object recognition, segmentation, etc. However, there is a technical issue with the design of the experiments of the study, and thus the correctness of the claim is yet to be verified. Moreover, the absolute image position may not be essential for the segmentation of natural images, in which target objects will appear at any image position. In this study, we investigate how positional information is and can be utilized for segmentation tasks. Toward this end, we consider {em positional encoding} (PE) that adds channels embedding image position to the input images and compare PE with several padding methods. Considering the above nature of natural images, we choose medical image segmentation tasks, in which the absolute position appears to be relatively important, as the same organs (of different patients) are captured in similar sizes and positions. We draw a mixed conclusion from the experimental results; the positional encoding certainly works in some cases, but the absolute image position may not be so important for segmentation tasks as we think.

We empirically evaluate an undervolting technique, i.e., underscaling the circuit supply voltage below the nominal level, to improve the power-efficiency of Convolutional Neural Network (CNN) accelerators mapped to Field Programmable Gate Arrays (FPGAs). Undervolting below a safe voltage level can lead to timing faults due to excessive circuit latency increase. We evaluate the reliability-power trade-off for such accelerators. Specifically, we experimentally study the reduced-voltage operation of multiple components of real FPGAs, characterize the corresponding reliability behavior of CNN accelerators, propose techniques to minimize the drawbacks of reduced-voltage operation, and combine undervolting with architectural CNN optimization techniques, i.e., quantization and pruning. We investigate the effect of environmental temperature on the reliability-power trade-off of such accelerators. We perform experiments on three identical samples of modern Xilinx ZCU102 FPGA platforms with five state-of-the-art image classification CNN benchmarks. This approach allows us to study the effects of our undervolting technique for both software and hardware variability. We achieve more than 3X power-efficiency (GOPs/W) gain via undervolting. 2.6X of this gain is the result of eliminating the voltage guardband region, i.e., the safe voltage region below the nominal level that is set by FPGA vendor to ensure correct functionality in worst-case environmental and circuit conditions. 43% of the power-efficiency gain is due to further undervolting below the guardband, which comes at the cost of accuracy loss in the CNN accelerator. We evaluate an effective frequency underscaling technique that prevents this accuracy loss, and find that it reduces the power-efficiency gain from 43% to 25%.

This paper reviews the second challenge on spectral reconstruction from RGB images, i.e., the recovery of whole-scene hyperspectral (HS) information from a 3-channel RGB image. As in the previous challenge, two tracks were provided: (i) a "Clean" track where HS images are estimated from noise-free RGBs, the RGB images are themselves calculated numerically using the ground-truth HS images and supplied spectral sensitivity functions (ii) a "Real World" track, simulating capture by an uncalibrated and unknown camera, where the HS images are recovered from noisy JPEG-compressed RGB images. A new, larger-than-ever, natural hyperspectral image data set is presented, containing a total of 510 HS images. The Clean and Real World tracks had 103 and 78 registered participants respectively, with 14 teams competing in the final testing phase. A description of the proposed methods, alongside their challenge scores and an extensive evaluation of top performing methods is also provided. They gauge the state-of-the-art in spectral reconstruction from an RGB image.

Low-resolution text images are often seen in natural scenes such as documents captured by mobile phones. Recognizing low-resolution text images is challenging because they lose detailed content information, leading to poor recognition accuracy. An intuitive solution is to introduce super-resolution (SR) techniques as pre-processing. However, previous single image super-resolution (SISR) methods are trained on synthetic low-resolution images (e.g.Bicubic down-sampling), which is simple and not suitable for real low-resolution text recognition. To this end, we pro-pose a real scene text SR dataset, termed TextZoom. It contains paired real low-resolution and high-resolution images which are captured by cameras with different focal length in the wild. It is more authentic and challenging than synthetic data, as shown in Fig. 1. We argue improv-ing the recognition accuracy is the ultimate goal for Scene Text SR. In this purpose, a new Text Super-Resolution Network termed TSRN, with three novel modules is developed. (1) A sequential residual block is proposed to extract the sequential information of the text images. (2) A boundary-aware loss is designed to sharpen the character boundaries. (3) A central alignment module is proposed to relieve the misalignment problem in TextZoom. Extensive experiments on TextZoom demonstrate that our TSRN largely improves the recognition accuracy by over 13%of CRNN, and by nearly 9.0% of ASTER and MORAN compared to synthetic SR data. Furthermore, our TSRN clearly outperforms 7 state-of-the-art SR methods in boosting the recognition accuracy of LR images in TextZoom. For example, it outperforms LapSRN by over 5% and 8%on the recognition accuracy of ASTER and CRNN. Our results suggest that low-resolution text recognition in the wild is far from being solved, thus more research effort is needed.

Convolutional Neural Networks (CNNs) are generally prone to noise interruptions, i.e., small image noise can cause drastic changes in the output. To suppress the noise effect to the final predication, we enhance CNNs by replacing max-pooling, strided-convolution, and average-pooling with Discrete Wavelet Transform (DWT). We present general DWT and Inverse DWT (IDWT) layers applicable to various wavelets like Haar, Daubechies, and Cohen, etc., and design wavelet integrated CNNs (WaveCNets) using these layers for image classification. In WaveCNets, feature maps are decomposed into the low-frequency and high-frequency components during the down-sampling. The low-frequency component stores main information including the basic object structures, which is transmitted into the subsequent layers to extract robust high-level features. The high-frequency components, containing most of the data noise, are dropped during inference to improve the noise-robustness of the WaveCNets. Our experimental results on ImageNet and ImageNet-C (the noisy version of ImageNet) show that WaveCNets, the wavelet integrated versions of VGG, ResNets, and DenseNet, achieve higher accuracy and better noise-robustness than their vanilla versions.

Most studies on speaker verification systems focus on long-duration utterances, which are composed of sufficient phonetic information. However, the performances of these systems are known to degrade when short-duration utterances are inputted due to the lack of phonetic information as compared to the long utterances. In this paper, we propose a method that compensates for the performance degradation of speaker verification for short utterances, referred to as "crop aggregating". The proposed method adopts an ensemble-based design to improve the stability and accuracy of speaker verification systems. The proposed method segments an input utterance into several short utterances and then aggregates the segment embeddings extracted from the segmented inputs to compose a speaker embedding. Then, this method simultaneously trains the segment embeddings and the aggregated speaker embedding. In addition, we also modified the teacher-student learning method for the proposed method. Experimental results on different input duration using the VoxCeleb1 test set demonstrate that the proposed technique improves speaker verification performance by about 45.37% relatively compared to the baseline system with 1-second test utterance condition.

Similarity measure, also called information measure, is a concept used to distinguish different objects. It has been studied from different contexts by employing mathematical, psychological, and fuzzy approaches. Image steganography is the art of hiding secret data into an image in such a way that it cannot be detected by an intruder. In image steganography, hiding secret data in the plain or non-edge regions of the image is significant due to the high similarity and redundancy of the pixels in their neighborhood. However, the similarity measure of the neighboring pixels, i.e., their proximity in color space, is perceptual rather than mathematical. This paper proposes an interval type 2 fuzzy logic system (IT2 FLS) to determine the similarity between the neighboring pixels by involving an instinctive human perception through a rule-based approach. The pixels of the image having high similarity values, calculated using the proposed IT2 FLS similarity measure, are selected for embedding via the least significant bit (LSB) method. We term the proposed procedure of steganography as IT2 FLS LSB method. Moreover, we have developed two more methods, namely, type 1 fuzzy logic system based least significant bits (T1FLS LSB) and Euclidean distance based similarity measures for least significant bit (SM LSB) steganographic methods. Experimental simulations were conducted for a collection of images and quality index metrics, such as PSNR, UQI, and SSIM are used. All the three steganographic methods are applied on datasets and the quality metrics are calculated. The obtained stego images and results are shown and thoroughly compared to determine the efficacy of the IT2 FLS LSB method. Finally, we have done a comparative analysis of the proposed approach with the existing well-known steganographic methods to show the effectiveness of our proposed steganographic method.

Visualization-oriented natural language interfaces (V-NLIs) have been explored and developed in recent years. One challenge faced by V-NLIs is in the formation of effective design decisions that usually requires a deep understanding of user queries. Learning-based approaches have shown potential in V-NLIs and reached state-of-the-art performance in various NLP tasks. However, because of the lack of sufficient training samples that cater to visual data analytics, cutting-edge techniques have rarely been employed to facilitate the development of V-NLIs. We present a new dataset, called Quda, to help V-NLIs understand free-form natural language. Our dataset contains 14;035 diverse user queries annotated with 10 low-level analytic tasks that assist in the deployment of state-of-the-art techniques for parsing complex human language. We achieve this goal by first gathering seed queries with data analysts who are target users of V-NLIs. Then we employ extensive crowd force for paraphrase generation and validation. We demonstrate the usefulness of Quda in building V-NLIs by creating a prototype that makes effective design decisions for free-form user queries. We also show that Quda can be beneficial for a wide range of applications in the visualization community by analyzing the design tasks described in academic publications.

A method for encoding information in DNA sequences is described. The method is based on the precision-resolution framework, and is aimed to work in conjunction with a recently suggested terminator-free template independent DNA synthesis method. The suggested method optimizes the amount of information bits per synthesis time unit, namely, the writing rate. Additionally, the encoding scheme studied here takes into account the existence of multiple copies of the DNA sequence, which are independently distorted. Finally, quantizers for various run-length distributions are designed.

Recently, the outbreak of Coronavirus Disease 2019 (COVID-19) has spread rapidly across the world. Due to the large number of affected patients and heavy labor for doctors, computer-aided diagnosis with machine learning algorithm is urgently needed, and could largely reduce the efforts of clinicians and accelerate the diagnosis process. Chest computed tomography (CT) has been recognized as an informative tool for diagnosis of the disease. In this study, we propose to conduct the diagnosis of COVID-19 with a series of features extracted from CT images. To fully explore multiple features describing CT images from different views, a unified latent representation is learned which can completely encode information from different aspects of features and is endowed with promising class structure for separability. Specifically, the completeness is guaranteed with a group of backward neural networks (each for one type of features), while by using class labels the representation is enforced to be compact within COVID-19/community-acquired pneumonia (CAP) and also a large margin is guaranteed between different types of pneumonia. In this way, our model can well avoid overfitting compared to the case of directly projecting highdimensional features into classes. Extensive experimental results show that the proposed method outperforms all comparison methods, and rather stable performances are observed when varying the numbers of training data.

In the future we can expect that artificial intelligent agents, once deployed, will be required to learn continually from their experience during their operational lifetime. Such agents will also need to communicate with humans and other agents regarding the content of their experience, in the context of passing along their learnings, for the purpose of explaining their actions in specific circumstances or simply to relate more naturally to humans concerning experiences the agent acquires that are not necessarily related to their assigned tasks. We argue that to support these goals, an agent would benefit from an episodic memory; that is, a memory that encodes the agent's experience in such a way that the agent can relive the experience, communicate about it and use its past experience, inclusive of the agents own past actions, to learn more effective models and policies. In this short paper, we propose one potential approach to provide an AI agent with such capabilities. We draw upon the ever-growing body of work examining the function and operation of the Medial Temporal Lobe (MTL) in mammals to guide us in adding an episodic memory capability to an AI agent composed of artificial neural networks (ANNs). Based on that, we highlight important aspects to be considered in the memory organization and we propose an architecture combining ANNs and standard Computer Science techniques for supporting storage and retrieval of episodic memories. Despite being initial work, we hope this short paper can spark discussions around the creation of intelligent agents with memory or, at least, provide a different point of view on the subject.

This paper reviews the Challenge on Image Demoireing that was part of the New Trends in Image Restoration and Enhancement (NTIRE) workshop, held in conjunction with CVPR 2020. Demoireing is a difficult task of removing moire patterns from an image to reveal an underlying clean image. The challenge was divided into two tracks. Track 1 targeted the single image demoireing problem, which seeks to remove moire patterns from a single image. Track 2 focused on the burst demoireing problem, where a set of degraded moire images of the same scene were provided as input, with the goal of producing a single demoired image as output. The methods were ranked in terms of their fidelity, measured using the peak signal-to-noise ratio (PSNR) between the ground truth clean images and the restored images produced by the participants' methods. The tracks had 142 and 99 registered participants, respectively, with a total of 14 and 6 submissions in the final testing stage. The entries span the current state-of-the-art in image and burst image demoireing problems.

We propose an augmented Lagrangian preconditioner for a three-field stress-velocity-pressure discretization of stationary non-Newtonian incompressible flow with an implicit constitutive relation of power-law type. The discretization employed makes use of the divergence-free Scott-Vogelius pair for the velocity and pressure. The preconditioner builds on the work [P. E. Farrell, L. Mitchell, and F. Wechsung, SIAM J. Sci. Comput., 41 (2019), pp. A3073-A3096], where a Reynolds-robust preconditioner for the three-dimensional Newtonian system was introduced. The preconditioner employs a specialized multigrid method for the stress-velocity block that involves a divergence-capturing space decomposition and a custom prolongation operator. The solver exhibits excellent robustness with respect to the parameters arising in the constitutive relation, allowing for the simulation of a wide range of materials.

Smart meters enable remote and automatic electricity, water and gas consumption reading and are being widely deployed in developed countries. Nonetheless, there is still a huge number of non-smart meters in operation. Image-based Automatic Meter Reading (AMR) focuses on dealing with this type of meter readings. We estimate that the Energy Company of Paran'a (Copel), in Brazil, performs more than 850,000 readings of dial meters per month. Those meters are the focus of this work. Our main contributions are: (i) a public real-world dial meter dataset (shared upon request) called UFPR-ADMR; (ii) a deep learning-based recognition baseline on the proposed dataset; and (iii) a detailed error analysis of the main issues present in AMR for dial meters. To the best of our knowledge, this is the first work to introduce deep learning approaches to multi-dial meter reading, and perform experiments on unconstrained images. We achieved a 100.0% F1-score on the dial detection stage with both Faster R-CNN and YOLO, while the recognition rates reached 93.6% for dials and 75.25% for meters using Faster R-CNN (ResNext-101).

With the growing reliance on the ubiquitous availability of IT systems and services, these systems become more global, scaled, and complex to operate. To maintain business viability, IT service providers must put in place reliable and cost efficient operations support. Artificial Intelligence for IT Operations (AIOps) is a promising technology for alleviating operational complexity of IT systems and services. AIOps platforms utilize big data, machine learning and other advanced analytics technologies to enhance IT operations with proactive actionable dynamic insight.

In this paper we share our experience applying the AIOps approach to a production cloud object storage service to get actionable insights into system's behavior and health. We describe a real-life production cloud scale service and its operational data, present the AIOps platform we have created, and show how it has helped us resolving operational pain points.

Linkage Tree Genetic Algorithm (LTGA) is an effective Evolutionary Algorithm (EA) to solve complex problems using the linkage information between problem variables. LTGA performs well in various kinds of single-task optimization and yields promising results in comparison with the canonical genetic algorithm. However, LTGA is an unsuitable method for dealing with multi-task optimization problems. On the other hand, Multifactorial Optimization (MFO) can simultaneously solve independent optimization problems, which are encoded in a unified representation to take advantage of the process of knowledge transfer. In this paper, we introduce Multifactorial Linkage Tree Genetic Algorithm (MF-LTGA) by combining the main features of both LTGA and MFO. MF-LTGA is able to tackle multiple optimization tasks at the same time, each task learns the dependency between problem variables from the shared representation. This knowledge serves to determine the high-quality partial solutions for supporting other tasks in exploring the search space. Moreover, MF-LTGA speeds up convergence because of knowledge transfer of relevant problems. We demonstrate the effectiveness of the proposed algorithm on two benchmark problems: Clustered Shortest-Path Tree Problem and Deceptive Trap Function. In comparison to LTGA and existing methods, MF-LTGA outperforms in quality of the solution or in computation time.