Related papers: CLAP. I. Resolving miscalibration for deep learnin…
Many of the cosmological tests to be performed by planned dark energy experiments will require extremely well-characterized photometric redshift measurements. Current estimates are that the true mean redshift of the objects in each photo-z…
Contrastive learning has emerged as a powerful method in deep learning, excelling at learning effective representations through contrasting samples from different distributions. However, neural collapse, where embeddings converge into a…
Existing vision-text contrastive learning models enhance representation transferability and support zero-shot prediction by matching paired image and caption embeddings while pushing unrelated pairs apart. However, astronomical image-label…
Photometric redshift estimation plays a crucial role in modern cosmological surveys for studying the universe's large-scale structures and the evolution of galaxies. Deep learning has emerged as a powerful method to produce accurate…
Calibrating deep learning models to yield uncertainty-aware predictions is crucial as deep neural networks get increasingly deployed in safety-critical applications. While existing post-hoc calibration methods achieve impressive results on…
This paper addresses the critical issue of miscalibration in CLIP-based model adaptation, particularly in the challenging scenario of out-of-distribution (OOD) samples, which has been overlooked in the existing literature on CLIP…
Solving point-wise feature correspondence in visual data is a fundamental problem in computer vision. A powerful model that addresses this challenge is to formulate it as graph matching, which entails solving a Quadratic Assignment Problem…
Multi-label image classification presents a challenging task in many domains, including computer vision and medical imaging. Recent advancements have introduced graph-based and transformer-based methods to improve performance and capture…
Convolutional neural networks (CNNs) have achieved superhuman performance in multiple vision tasks, especially image classification. However, unlike humans, CNNs leverage spurious features, such as background information to make decisions.…
In recent years, deep neural networks (DNNs) have shown competitive results in many fields. Despite this success, they often suffer from poor calibration, especially in safety-critical scenarios such as autonomous driving and healthcare,…
We present results exploring the role that probabilistic deep learning models can play in cosmology from large scale astronomical surveys through estimating the distances to galaxies (redshifts) from photometry. Due to the massive scale of…
Photometric redshifts (photo-z's) provide an alternative way to estimate the distances of large samples of galaxies and are therefore crucial to a large variety of cosmological problems. Among the various methods proposed over the years,…
We present results exploring the role that probabilistic deep learning models can play in cosmology from large-scale astronomical surveys through photometric redshift (photo-z) estimation. Photo-z uncertainty estimates are critical for the…
Many contrastive representation learning methods learn a single global representation of an entire image. However, dense contrastive representation learning methods such as DenseCL (Wang et al., 2021) can learn better representations for…
Accurate photometric redshifts are among the key requirements for precision weak lensing measurements. Both the large size of the Sloan Digital Sky Survey (SDSS) and the existence of large spectroscopic redshift samples that are…
Precision cosmology requires accurate galaxy redshifts, but next generation optical surveys will observe unprecedented numbers of resolved galaxies, placing strain on the amount of spectroscopic follow-up required. We show how useful…
Calibrating the redshift distributions of photometric galaxy samples is essential in weak lensing studies. The self-calibration method combines angular auto- and cross-correlations between galaxies in multiple photometric redshift…
Learning with noisy labels (LNL) has been extensively studied, with existing approaches typically following a framework that alternates between clean sample selection and semi-supervised learning (SSL). However, this approach has a…
Cross-correlations between the galaxy number density in a lensing source sample and that in an overlapping spectroscopic sample can in principle be used to calibrate the lensing source redshift distribution. In this paper, we study in…
We present a new machine learning model for estimating photometric redshifts with improved accuracy for galaxies in Pan-STARRS1 data release 1. Depending on the estimation range of redshifts, this model based on neural networks can handle…