Related papers: Elf autoencoder: unsupervised exploration of flat-…
Enhancing superconductivity through material design is a central goal in quantum materials research. Moire engineering, where twisting stacked layers creates long-wavelength modulations and flat bands, has shown how electronic correlations…
The modification of electronic band structures and the subsequent tuning of electrical, optical, and thermal material properties is a central theme in the engineering and fundamental understanding of solid-state systems. In this scenario,…
We introduce Equivariant Neural Field Expectation Maximization (EFEM), a simple, effective, and robust geometric algorithm that can segment objects in 3D scenes without annotations or training on scenes. We achieve such unsupervised…
Text autoencoders are often used for unsupervised conditional text generation by applying mappings in the latent space to change attributes to the desired values. Recently, Mai et al. (2020) proposed Emb2Emb, a method to learn these…
Unsupervised learning is becoming more and more important recently. As one of its key components, the autoencoder (AE) aims to learn a latent feature representation of data which is more robust and discriminative. However, most AE based…
In this paper we propose Structuring AutoEncoders (SAE). SAEs are neural networks which learn a low dimensional representation of data which are additionally enriched with a desired structure in this low dimensional space. While traditional…
Federated learning (FL) has been recognized as one of the most promising solutions for distributed machine learning (ML). In most of the current literature, FL has been studied for supervised ML tasks, in which edge devices collect labeled…
The discovery of two-dimensional (2D) magnetism within atomically thin structures derived from layered crystals has opened up a new realm for exploring magnetic heterostructures. This emerging field provides a foundational platform for…
In this work, we introduce LEAD, an approach to discover landmarks from an unannotated collection of category-specific images. Existing works in self-supervised landmark detection are based on learning dense (pixel-level) feature…
The full-potential linearized augmented plane wave method with the generalized gradient approximation for the exchange-correlation potential (FLAPW-GGA) is used to predict the electronic and elastic properties of the newly discovered…
Flat bands, in which kinetic energy is quenched and quantum states become macroscopically degenerate, host a rich variety of correlated and topological phases, from unconventional superconductors to fractional Chern insulators. In Hermitian…
We propose the Autoencoding Binary Classifiers (ABC), a novel supervised anomaly detector based on the Autoencoder (AE). There are two main approaches in anomaly detection: supervised and unsupervised. The supervised approach accurately…
Electronic correlations in two-dimensional materials play a crucial role in stabilising emergent phases of matter. The realisation of correlation-driven phenomena in graphene has remained a longstanding goal, primarily due to the absence of…
Even though the term `organic electronics' evokes rather organic devices, a significant part of its scope deals with physical properties of `active elements' such as organic films and interfaces. Examination of the film growth and the…
Sparse autoencoders are a standard tool for uncovering interpretable latent representations in neural networks. Yet, their interpretation depends on the inputs, making their isolated study incomplete. Polynomials offer a solution; they…
Paper surfaces under the microscopic view are observed to be formed by intertwisted wood fibers. Such structures of paper surfaces are unique from one location to another and are almost impossible to duplicate. Previous work used…
Photonic flat bands are crucial for enabling strong localization of light and enhancing light-matter interactions, as well as tailoring the angular distribution of emission from photonic structures. These unique properties open pathways for…
We study characteristic electronic structures in an extended martini lattice model and propose its materialization in $\pi$-electron networks constructed by designated chemisorption on graphene and silicene. By investigating the minimal…
It has long been noticed that special lattices contain single-electron flat bands (FB) without any dispersion. Since the kinetic energy of electrons is quenched in the FB, this highly degenerate energy level becomes an ideal platform to…
Raman spectroscopy is widely used across scientific domains to characterize the chemical composition of samples in a non-destructive, label-free manner. Many applications entail the unmixing of signals from mixtures of molecular species to…