Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userInsights into one-body density matrices using deep learning
92
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
The one-body reduced density matrix (1-RDM) of a many-body system at zero temperature gives direct access to many observables, such as the charge density, kinetic energy and occupation numbers. It would be desirable to express it as a simple functional of the density or of other local observables, but to date satisfactory approximations have not yet been found. Deep learning is the state-of the art approach to perform high dimensional regressions and classification tasks, and is becoming widely used in the condensed matter community to develop increasingly accurate density functionals. Autoencoders are deep learning models that perform efficient dimensionality reduction, allowing the distillation of data to its fundamental features needed to represent it. By training autoencoders on a large data-set of 1-RDMs from exactly solvable real-space model systems, and performing principal component analysis, the machine learns to what extent the data can be compressed and hence how it is constrained. We gain insight into these machine learned constraints and employ them to inform approximations to the 1-RDM as a functional of the charge density. We exploit known physical properties of the 1-RDM in the simplest possible cases to perform feature engineering, where we inform the structure of the models from known mathematical relations, allowing us to integrate existing understanding into the machine learning methods. By comparing various deep learning approaches we gain insight into what physical features of the density matrix are most amenable to machine learning, utilising both known and learned characteristics.
rate research
Read More
While the evolution of linear initial conditions present in the early universe into extended halos of dark matter at late times can be computed using cosmological simulations, a theoretical understanding of this complex process remains elusive. Here, we build a deep learning framework to learn this non-linear relationship, and develop techniques to physically interpret the learnt mapping. A three-dimensional convolutional neural network (CNN) is trained to predict the mass of dark matter halos from the initial conditions. We find no change in the predictive accuracy of the model if we retrain the model removing anisotropic information from the inputs. This suggests that the features learnt by the CNN are equivalent to spherical averages over the initial conditions. Our results indicate that interpretable deep learning frameworks can provide a powerful tool for extracting insight into cosmological structure formation.
Due to non-linear structure, iterative Greens function methods can result in multiple different solutions even for simple molecular systems. In contrast to the wave-function methods, a detailed and careful analysis of such molecular solutions was not performed before. In this work, we use two-particle density matrices to investigate local spin and charge correlators that quantify the charge-resonance and covalent characters of these solutions. When applied within unrestricted orbital set, spin correlators elucidate the broken symmetry of the solutions, containing necessary information for building effective magnetic Hamiltonians. Based on GW and GF2 calculations of simple molecules and transition metal complexes, we construct Heisenberg Hamiltonians, four-spin-four-center corrections, as well as biquadratic spin-spin interactions. These Hamiltonian parametrizations are compared to prior wave-function calculations.
We explore the application of computer vision and machine learning (ML) techniques to predict material properties (e.g. compressive strength) based on SEM images. We show that its possible to train ML models to predict materials performance based on SEM images alone, demonstrating this capability on the real-world problem of predicting uniaxially compressed peak stress of consolidated molecular solids samples. Our image-based ML approach reduces mean absolute percent error (MAPE) by an average of 24% over baselines representative of the current state-of-the-practice (i.e., domain-experts analysis and correlation). We compared two complementary approaches to this problem: (1) a traditional ML approach, random forest (RF), using state-of-the-art computer vision features and (2) an end-to-end deep learning (DL) approach, where features are learned automatically from raw images. We demonstrate the complementarity of these approaches, showing that RF performs best in the small data regime in which many real-world scientific applications reside (up to 24% lower RMSE than DL), whereas DL outpaces RF in the big data regime, where abundant training samples are available (up to 24% lower RMSE than RF). Finally, we demonstrate that models trained using machine learning techniques are capable of discovering and utilizing informative crystal attributes previously underutilized by domain experts.
Many applications in computational science require computing the elements of a function of a large matrix. A commonly used approach is based on the the evaluation of the eigenvalue decomposition, a task that, in general, involves a computing time that scales with the cube of the size of the matrix. We present here a method that can be used to evaluate the elements of a function of a positive-definite matrix with a scaling that is linear for sparse matrices and quadratic in the general case. This methodology is based on the properties of the dynamics of a multidimensional harmonic potential coupled with colored noise generalized Langevin equation (GLE) thermostats. This $f-$thermostat (FTH) approach allows us to calculate directly elements of functions of a positive-definite matrix by carefully tailoring the properties of the stochastic dynamics. We demonstrate the scaling and the accuracy of this approach for both dense and sparse problems and compare the results with other established methodologies.
In this paper, we study the importance of pruning in Deep Networks (DNs) and the yin & yang relationship between (1) pruning highly overparametrized DNs that have been trained from random initialization and (2) training small DNs that have been cleverly initialized. As in most cases practitioners can only resort to random initialization, there is a strong need to develop a grounded understanding of DN pruning. Current literature remains largely empirical, lacking a theoretical understanding of how pruning affects DNs decision boundary, how to interpret pruning, and how to design corresponding principled pruning techniques. To tackle those questions, we propose to employ recent advances in the theoretical analysis of Continuous Piecewise Affine (CPA) DNs. From this perspective, we will be able to detect the early-bird (EB) ticket phenomenon, provide interpretability into current pruning techniques, and develop a principled pruning strategy. In each step of our study, we conduct extensive experiments supporting our claims and results; while our main goal is to enhance the current understanding towards DN pruning instead of developing a new pruning method, our spline pruning criteria in terms of layerwise and global pruning is on par with or even outperforms state-of-the-art pruning methods.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
