Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userLearning neutrino effects in Cosmology with Convolutional Neural Networks
76
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
Measuring the sum of the three active neutrino masses, $M_ u$, is one of the most important challenges in modern cosmology. Massive neutrinos imprint characteristic signatures on several cosmological observables in particular on the large-scale structure of the Universe. In order to maximize the information that can be retrieved from galaxy surveys, accurate theoretical predictions in the non-linear regime are needed. Currently, one way to achieve those predictions is by running cosmological numerical simulations. Unfortunately, producing those simulations requires high computational resources -- seven hundred CPU hours for each neutrino mass case. In this work, we propose a new method, based on a deep learning network (U-Net), to quickly generate simulations with massive neutrinos from standard $Lambda$CDM simulations without neutrinos. We computed multiple relevant statistical measures of deep-learning generated simulations, and conclude that our method accurately reproduces the 3-dimensional spatial distribution of matter down to non-linear scales: $k < 0.7$ h/Mpc. Finally, our method allows us to generate massive neutrino simulations 10,000 times faster than the traditional methods.
rate research
Read More
Cosmological simulations play an important role in the interpretation of astronomical data, in particular in comparing observed data to our theoretical expectations. However, to compare data with these simulations, the simulations in principle need to include gravity, magneto-hydrodyanmics, radiative transfer, etc. These ideal large-volume simulations (gravo-magneto-hydrodynamical) are incredibly computationally expensive which can cost tens of millions of CPU hours to run. In this paper, we propose a deep learning approach to map from the dark-matter-only simulation (computationally cheaper) to the galaxy distribution (from the much costlier cosmological simulation). The main challenge of this task is the high sparsity in the target galaxy distribution: space is mainly empty. We propose a cascade architecture composed of a classification filter followed by a regression procedure. We show that our result outperforms a state-of-the-art model used in the astronomical community, and provides a good trade-off between computational cost and prediction accuracy.
The combination of current large scale structure and cosmic microwave background (CMB) anisotropies data can place strong constraints on the sum of the neutrino masses. Here we show that future cosmic shear experiments, in combination with CMB constraints, can provide the statistical accuracy required to answer questions about differences in the mass of individual neutrino species. Allowing for the possibility that masses are non-degenerate we combine Fisher matrix forecasts for a weak lensing survey like Euclid with those for the forthcoming Planck experiment. Under the assumption that neutrino mass splitting is described by a normal hierarchy we find that the combination Planck and Euclid will possibly reach enough sensitivity to put a constraint on the mass of a single species. Using a Bayesian evidence calculation we find that such future experiments could provide strong evidence for either a normal or an inverted neutrino hierachy. Finally we show that if a particular neutrino hierachy is assumed then this could bias cosmological parameter constraints, for example the dark energy equation of state parameter, by > 1sigma, and the sum of masses by 2.3sigma.
Convolutional neural networks trained without supervision come close to matching performance with supervised pre-training, but sometimes at the cost of an even higher number of parameters. Extracting subnetworks from these large unsupervised convnets with preserved performance is of particular interest to make them less computationally intensive. Typical pruning methods operate during training on a task while trying to maintain the performance of the pruned network on the same task. However, in self-supervised feature learning, the training objective is agnostic on the representation transferability to downstream tasks. Thus, preserving performance for this objective does not ensure that the pruned subnetwork remains effective for solving downstream tasks. In this work, we investigate the use of standard pruning methods, developed primarily for supervised learning, for networks trained without labels (i.e. on self-supervised tasks). We show that pruned masks obtained with or without labels reach comparable performance when re-trained on labels, suggesting that pruning operates similarly for self-supervised and supervised learning. Interestingly, we also find that pruning preserves the transfer performance of self-supervised subnetwork representations.
A machine learning technique with two-dimension convolutional neural network is proposed for detecting exoplanet transits. To test this new method, five different types of deep learning models with or without folding are constructed and studied. The light curves of the Kepler Data Release 25 are employed as the input of these models. The accuracy, reliability, and completeness are determined and their performances are compared. These results indicate that a combination of two-dimension convolutional neural network with folding would be an excellent choice for the future transit analysis.
We propose contextual convolution (CoConv) for visual recognition. CoConv is a direct replacement of the standard convolution, which is the core component of convolutional neural networks. CoConv is implicitly equipped with the capability of incorporating contextual information while maintaining a similar number of parameters and computational cost compared to the standard convolution. CoConv is inspired by neuroscience studies indicating that (i) neurons, even from the primary visual cortex (V1 area), are involved in detection of contextual cues and that (ii) the activity of a visual neuron can be influenced by the stimuli placed entirely outside of its theoretical receptive field. On the one hand, we integrate CoConv in the widely-used residual networks and show improved recognition performance over baselines on the core tasks and benchmarks for visual recognition, namely image classification on the ImageNet data set and object detection on the MS COCO data set. On the other hand, we introduce CoConv in the generator of a state-of-the-art Generative Adversarial Network, showing improved generative results on CIFAR-10 and CelebA. Our code is available at https://github.com/iduta/coconv.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
