Discovering the underlying behavior of complex systems is an important topic in many science and engineering disciplines. In this paper, we propose a novel neural network framework, finite difference neural networks (FD-Net), to learn partial differential equations from data. Specifically, our proposed finite difference inspired network is designed to learn the underlying governing partial differential equations from trajectory data, and to iteratively estimate the future dynamical behavior using only a few trainable parameters. We illustrate the performance (predictive power) of our framework on the heat equation, with and without noise and/or forcing, and compare our results to the Forward Euler method. Moreover, we show the advantages of using a Hessian-Free Trust Region method to train the network.

Advanced data augmentation strategies have widely been studied to improve the generalization ability of deep learning models. Regional dropout is one of the popular solutions that guides the model to focus on less discriminative parts by randomly removing image regions, resulting in improved regularization. However, such information removal is undesirable. On the other hand, recent strategies suggest to randomly cut and mix patches and their labels among training images, to enjoy the advantages of regional dropout without having any pointless pixel in the augmented images. We argue that the random selection of the patch may not necessarily represent any information about the corresponding object and thereby mixing the labels according to that uninformative patch enables the model to learn unexpected feature representation. Therefore, we propose SaliencyMix that carefully selects a representative image patch with the help of a saliency map and mixes this indicative patch with the target image that leads the model to learn more appropriate feature representation. SaliencyMix achieves a new state-of-the-art top-1 error of 20.09% on ImageNet classification using ResNet-101 architecture and also improves the model robustness against adversarial perturbations. Furthermore, SaliencyMix trained model helps to improve the object detection performance.

The data-centric construction of inexpensive surrogates for fine-grained, physical models has been at the forefront of computational physics due to its significant utility in many-query tasks such as uncertainty quantification. Recent efforts have taken advantage of the enabling technologies from the field of machine learning (e.g. deep neural networks) in combination with simulation data. While such strategies have shown promise even in higher-dimensional problems, they generally require large amounts of training data even though the construction of surrogates is by definition a Small Data problem. Rather than employing data-based loss functions, it has been proposed to make use of the governing equations (in the simplest case at collocation points) in order to imbue domain knowledge in the training of the otherwise black-box-like interpolators. The present paper provides a flexible, probabilistic framework that accounts for physical structure and information both in the training objectives as well as in the surrogate model itself. We advocate a probabilistic (Bayesian) model in which equalities that are available from the physics (e.g. residuals, conservation laws) can be introduced as virtual observables and can provide additional information through the likelihood. We further advocate a generative model i.e. one that attempts to learn the joint density of inputs and outputs that is capable of making use of unlabeled data (i.e. only inputs) in a semi-supervised fashion in order to promote the discovery of lower-dimensional embeddings which are nevertheless predictive of the fine-grained model's output.

Recent work on exploration in reinforcement learning (RL) has led to a series of increasingly complex solutions to the problem. This increase in complexity often comes at the expense of generality. Recent empirical studies suggest that, when applied to a broader set of domains, some sophisticated exploration methods are outperformed by simpler counterparts, such as {\epsilon}-greedy. In this paper we propose an exploration algorithm that retains the simplicity of {\epsilon}-greedy while reducing dithering. We build on a simple hypothesis: the main limitation of {\epsilon}-greedy exploration is its lack of temporal persistence, which limits its ability to escape local optima. We propose a temporally extended form of {\epsilon}-greedy that simply repeats the sampled action for a random duration. It turns out that, for many duration distributions, this suffices to improve exploration on a large set of domains. Interestingly, a class of distributions inspired by ecological models of animal foraging behaviour yields particularly strong performance.

In the artificial neuron, I replace the dot product with the weighted Lehmer mean, which may emulate different cases of a generalized mean. The single neuron instance is replaced by a multiplet of neurons which have the same averaging weights. A group of outputs feed forward, in lieu of the single scalar. The generalization parameter is typically set to a different value for each neuron in the multiplet. I further extend the concept to a multiplet taken from the Gini mean. Derivatives with respect to the weight parameters and with respect to the two generalization parameters are given. Some properties of the network are investigated, showing the capacity to emulate the classical exclusive-or problem organically in two layers and perform some multiplication and division. The network can instantiate truncated power series and variants, which can be used to approximate different functions, provided that parameters are constrained. Moreover, a mean case slope score is derived that can facilitate a learning-rate novelty based on homogeneity of the selected elements. The multiplet neuron equation provides a way to segment regularization timeframes and approaches.

The important peaks related to the physical properties of a lithium ion rechargeable battery were extracted from the measured X ray diffraction spectrum by a convolutional neural network based on the Attention mechanism. Among the deep features, the lattice constant of the cathodic active material was selected as a cell voltage predictor, and the crystallographic behavior of the active anodic and cathodic materials revealed the rate property during the charge discharge states. The machine learning automatically selected the significant peaks from the experimental spectrum. Applying the Attention mechanism with appropriate objective variables in multi task trained models, one can selectively visualize the correlations between interesting physical properties. As the deep features are automatically defined, this approach can adapt to the conditions of various physical experiments.

Deep convolution-based single image super-resolution (SISR) networks embrace the benefits of learning from large-scale external image resources for local recovery, yet most existing works have ignored the long-range feature-wise similarities in natural images. Some recent works have successfully leveraged this intrinsic feature correlation by exploring non-local attention modules. However, none of the current deep models have studied another inherent property of images: cross-scale feature correlation. In this paper, we propose the first Cross-Scale Non-Local (CS-NL) attention module with integration into a recurrent neural network. By combining the new CS-NL prior with local and in-scale non-local priors in a powerful recurrent fusion cell, we can find more cross-scale feature correlations within a single low-resolution (LR) image. The performance of SISR is significantly improved by exhaustively integrating all possible priors. Extensive experiments demonstrate the effectiveness of the proposed CS-NL module by setting new state-of-the-arts on multiple SISR benchmarks.

Deep neural networks are learning models having achieved state of the art performance in many fields like prediction, computer vision, language processing and so on. However, it has been shown that certain inputs exist which would not trick a human normally, but may mislead the model completely. These inputs are known as adversarial inputs. These inputs pose a high security threat when such models are used in real world applications. In this work, we have analyzed the resistance of three different classes of fully connected dense networks against the rarely tested non-gradient based adversarial attacks. These classes are created by manipulating the input and output layers. We have proven empirically that owing to certain characteristics of the network, they provide a high robustness against these attacks, and can be used in fine tuning other models to increase defense against adversarial attacks.

Measurement of the red damping wing of neutral hydrogen in quasar spectra provides a probe of the epoch of reionization in the early Universe. Such quantification requires precise and unbiased estimates of the intrinsic continua near Lyman-$\alpha$ (Ly$\alpha$), a challenging task given the highly variable Ly$\alpha$ emission profiles of quasars. Here, we introduce a fully probabilistic approach to intrinsic continua prediction. We frame the problem as a conditional density estimation task and explicitly model the distribution over plausible blue-side continua ($1190\ \unicode{xC5} \leq \lambda_{\text{rest}} < 1290\ \unicode{xC5}$) conditional on the red-side spectrum ($1290\ \unicode{xC5} \leq \lambda_{\text{rest}} < 2900\ \unicode{xC5}$) using normalizing flows. Our approach achieves state-of-the-art precision and accuracy, allows for sampling one thousand plausible continua in less than a tenth of a second, and can natively provide confidence intervals on the blue-side continua via Monte Carlo sampling. We measure the damping wing effect in two $z>7$ quasars and estimate the volume-averaged neutral fraction of hydrogen from each, finding $\bar{x}_\text{HI}=0.304 \pm 0.042$ for ULAS J1120+0641 ($z=7.09$) and $\bar{x}_\text{HI}=0.384 \pm 0.133$ for ULAS J1342+0928 ($z=7.54$).

The mechanism of message passing in graph neural networks(GNNs) is still mysterious for the literature. No one, to our knowledge, has given another possible theoretical origin for GNNs apart from convolutional neural networks. Somewhat to our surprise, the message passing can be best understood in terms of the power iteration. By removing activation functions and layer weights of GNNs, we propose power iteration clustering (SPIC) models which are naturally interpretable and scalable. The experiment shows our models extend the existing GNNs and enhance its capability of processing random featured networks. Moreover, we demonstrate the redundancy of some state-of-the-art GNNs in designing and define a lower limit for model evaluation by randomly initializing the aggregator of message passing. All the findings in this paper push the boundaries of our understanding of neural networks.