Related papers: A Probabilistic Autoencoder for Galaxy SED Reconst…

The cosmological exploitation of modern photometric galaxy surveys requires both accurate (unbiased) and precise (narrow) redshift probability distributions derived from broadband photometry. Existing methodologies do not meet those…

With the advent of billion-galaxy surveys with complex data, the need of the hour is to efficiently model galaxy spectral energy distributions (SEDs) with robust uncertainty quantification. The combination of Simulation-Based inference…

Sparse autoencoders (SAEs) have recently emerged as a powerful tool for interpreting the features learned by large language models (LLMs). By reconstructing features with sparsely activated networks, SAEs aim to recover complex superposed…

Variational autoencoders (VAE) often use Gaussian or category distribution to model the inference process. This puts a limit on variational learning because this simplified assumption does not match the true posterior distribution, which is…

The growth in the number of galaxy images is much faster than the speed at which these galaxies can be labelled by humans. However, by leveraging the information present in the ever growing set of unlabelled images, semi-supervised learning…

As a concrete setting where stochastic partial differential equations (SPDEs) are able to model real phenomena, we propose a stochastic Meinhardt model for cell repolarisation and study how parameter estimation techniques developed for…

Masked autoencoder (MAE), a simple and effective self-supervised learning framework based on the reconstruction of masked image regions, has recently achieved prominent success in a variety of vision tasks. Despite the emergence of…

Sparse autoencoders (SAEs) are a recent technique for decomposing neural network activations into human-interpretable features. However, in order for SAEs to identify all features represented in frontier models, it will be necessary to…

A new method is developed for estimating photometric redshifts, using realistic template SEDs, extending over four decades in wavelength (i.e from 0.05 $\mu m$ to 1 mm). The SEDs are constructed for four different spectral types of galaxies…

Sparse autoencoders (SAEs) have proven useful in disentangling the opaque activations of neural networks, primarily large language models, into sets of interpretable features. However, adapting them to domains beyond language, such as…

Realistic physical phenomena exhibit random fluctuations across many scales in the input and output processes. Models of these phenomena require stochastic PDEs. For three-dimensional coupled (vector-valued) stochastic PDEs (SPDEs), for…

Sparse autoencoders (SAEs) emerged as a promising tool for mechanistic interpretability of transformer-based foundation models. Very recently, SAEs were also adopted for the visual domain, enabling the discovery of visual concepts and their…

This letter proposes a design of low peak-to-average power ratio (PAPR), low symbol error rate (SER), and high data rate signal for optical orthogonal frequency division multiplexing (OFDM) systems. The proposed design leverages a…

We present PopSED, a framework for the population-level inference of galaxy properties from photometric data. Unlike the traditional approach of first analyzing individual galaxies and then combining the results to determine the physical…

The recently initiated SPHEREx and 7DS surveys will deliver low-resolution spectra ($R\approx 30-130$) for hundreds of millions of galaxies over the optical to near-infrared range ($0.4-5.0\mu m$), covering a wide sky area without sample…

Anomaly detection without priors of the anomalies is challenging. In the field of unsupervised anomaly detection, traditional auto-encoder (AE) tends to fail based on the assumption that by training only on normal images, the model will not…

Sparse Autoencoders (SAEs) have emerged as a powerful framework for machine learning interpretability, enabling the unsupervised decomposition of model representations into a dictionary of abstract, human-interpretable concepts. However, we…

Sparse Autoencoders (SAEs) are a powerful dictionary learning technique for decomposing neural network activations, translating the hidden state into human ideas with high semantic value despite no external intervention or guidance.…

Variational autoencoder (VAE) is a very successful generative model whose key element is the so called amortized inference network, which can perform test time inference using a single feed forward pass. Unfortunately, this comes at the…

Sparse autoencoders (SAEs) are widely used to extract human-interpretable features from neural network activations, but their learned features can vary substantially across random seeds and training choices. To improve stability, we studied…