Upcoming and Previous Seminars (Past months or Previous years)

Jan 16, Tuesday
1:10 pm (BCCP/Cosmology seminar)
Amelie Tamone, LBL
Campbell 131
"Improving BAO constraints with cosmic voids''
Cosmic voids defined as a subset of Delaunay Triangulation (DT) circumspheres have been used to measure the Baryon Acoustic Oscillations (BAO) scale. These voids are defined as spheres larger than a given radius threshold, which is constant over the survey volume. It was shown that they can help provide tighter constraints on cosmological parameters when combined with matter tracers. During this talk, I will present the latest results where we reanalysed the isotropic BAO measurements of different SDSS tracers combined with voids.

Jan 23, Tuesday
1:10 pm (BCCP/Cosmology seminar)
POSTPONED, no talk,
Campbell 131

Jan 25, Thursday
4 pm (LBL RPM)
Phil Harris,MIT
LBL 50-5132 and online
You can observe a lot by just watching (new AI strategies for the LHC and beyond)
With large amounts of data, a Higgs boson discovery, and world-leading constraints on an enormous amount of parameters and interactions, the Large Hadron Collider has been a phenomenal tool. However, it has reached the middle of life and is going through a mid-life crisis. We venture in a new direction to show how AI approaches are allowing us to do new, unprecedented physics measurements, leading to a mysterious deviation in Higgs boson production at high momentum. Moreover, we show a new result built on contrastive learning and semi-supervised learning strategies where, through physics-motivated choices, we teach an AI to visualize many physics processes simultaneously, allowing it to solve a variety of downstream tasks in one algorithm. The implications are far-reaching, including a world-leading Higgs boson identification and new sensitive approaches in AI-based anomaly detection. Finally, we show a dedicated path for the broad adoption of AI into all aspects of data processing at the Large Hadron Collider and other experiments, including LIGO, that are leading to faster, better, unprecedented results; the kind of results we believe can revitalize an aging accelerator.

Jan 26, Thursday
12 noon (INPA talk)
Bharath Nagam, Groningen
LBL 50A-5132 and ZOOM
Finding strong lenses using deep learning with upcoming Euclid data
Detecting strong lenses in a large dataset such as Euclid is very challenging due to the unbalanced nature of the dataset. Existing CNN models are producing large amounts of false positives, for example one stronglens candidate will be accompanied by 100's of false positives in the final sample. To overcome this challenge, we have developed a novel ML pipeline called DenseLens, which consists of three components namely Classification ensemble, Regression ensemble and Segmentation. Classification ensemble is an ensemble of DenseNet-CNNs which provides predictions in range [0,1] and Regression ensemble rank-orders strong lenses based on Information Content i.e., higher the rank, the more visually convincing features. Finally we use the segmentation model to predict the source pixels of the rank-ordered image. We use this additional information from this predicted source pixels to classify whether the candidate is a strong lens or not. We applied this novel approach of combining different ML models to the Kilo Degree Survey (KiDS) data and we reduced the false positives by an enormous factor. https://academic.oup.com/mnras/article-abstract/523/3/4188/7188322

Jan 30, Tuesday
CANCELLED due to illness
Georgios Valogiannis, U Chicago
Campbell 131
Precise Cosmological Constraints from BOSS Galaxy Clustering using the Wavelet Scattering Transform

Feb 6, Tuesday
1:10 pm (BCCP/Cosmology seminar)
Jamie Sullivan, UCB
Campbell 131
Galaxies Remember Inflation - New Aspects of Local Primordial non-Gaussianity in Galaxy Surveys
A signal of Local Primordial non-Gaussianity (LPNG) in the cosmological density field heralds the presence of new physics in the inflationary era. Optical and NIR galaxy surveys will soon exceed the sensitivity of the Cosmic Microwave Background and are posed to become the most powerful probe of such a signal. I will present recent and current work on the connection between galaxies tracing large-scale structure and LPNG in the early universe. Specifically, I will detail how splitting a population of galaxies in a survey by their clustering properties can increase the survey sensitivity to LPNG if host halo information beyond mass is considered. I will also discuss a new technique for estimating the response of galaxy number density to the presence of LPNG (the LPNG bias) directly from survey data. Finally, I will show the results of a consistency test of the large-scale perturbative biasing framework at the field level in the context of LPNG.

Feb 13, Tuesday
1:10 pm (BCCP/Cosmology seminar)
Kirit Karkare, SLAC
Campbell 131
Exploring the High-Redshift Universe with SPT-SLIM and Millimeter-Wave Line Intensity Mapping
Answering outstanding questions in cosmology - such as understanding the nature of inflation, dark energy, and reionization - requires observations of ever-increasing volumes of the universe. In this talk I will discuss a new technique for measuring large volumes at high redshift: line intensity mapping (LIM) of far-IR emission lines. This technique is enabled by advances in millimeter-wave spectrometer technology. I will introduce SPT-SLIM, a pathfinder experiment at the South Pole Telescope that will demonstrate the use of superconducting on-chip spectrometers for LIM. SPT-SLIM targets CO-emitting galaxies from 0.5 < z < 2. I will then discuss the future of mm-wave LIM and the technical advances needed to develop this technique into a next-generation cosmological observable.

Feb 15, Thursday
4pm LBL (RPM)
Will Coulton, Cambridge
LBL 50A-5132 (Sessler room) and ZOOM
Novel cosmological tests with precision CMB measurements
High resolution cosmic microwave background (CMB) experiments provide views of the Universe both at redshift ~1100, when the primary CMB was produced, as well as an integrated view of the Universe between then and now, through CMB secondary anisotropies. In this talk, I will show that CMB secondary anisotropies contain a wealth of cosmological information and are becoming leading cosmological probes. I will present recent measurements from the Atacama Cosmology Telescope collaboration that have mapped out several of these effects. I will then highlight some challenges for these measurements with upcoming CMB experiments, and some unique opportunities, such as tests of the fundamental properties of our Universe.

Feb 20, Tuesday
1:10 pm (BCCP/Cosmology seminar)
Minas Karamanis, UCB
Campbell 131
Preconditioned Monte Carlo: A New Tool for Bayesian Inference in Cosmology
Bayesian inference is paramount to cosmological analyses but relies on efficient numerical methods to explore high-dimensional, nonlinear posteriors. Current and forthcoming datasets now pose unprecedented challenges, with large parameter spaces, computational bottlenecks, and sparse data. Traditional sampling techniques struggle or fail in this regime, presenting a crisis and opportunity for new scalable algorithms. We introduce Preconditioned Monte Carlo (PMC), a novel algorithm leveraging geometric insights to enable robust and efficient sampling. We demonstrate empirically that PMC provides substantially improved posterior sampling efficiency and evidence estimation while favorably scaling to high dimensions. PMC offers a promising solution for Bayesian inference in cosmology. We discuss extensions and connections to related methods, and introduce pocoMC, our open-source PMC implementation for the cosmology community.

Feb 20, Tuesday
4pm LBL (RPM)
Clara Verges, Harvard
LBL 50A-5132 (Sessler room) and ZOOM
A new era for cosmology with current and next-generation CMB experiments
Cosmology has experienced a considerable surge in the past thirty years, and is now well established as a precision science. While the standard cosmological model provides an effective description of the observed Universe, many fundamental questions remain unresolved. From shedding light on the very first fractions of seconds after the Big Bang to looking for exotic physics, the Cosmic Microwave Background continues to be a unique probe for fundamental physics. The next generation of CMB polarisation observatories, embodied by CMB-Stage 4 for ground observations, will explore this window into the Universe in an unprecedented way, enabling breakthrough science. I will review how current experiments such as the BICEP/Keck telescopes pave the way for this next stage, by developing efficient technologies and observing strategies, and by sharpening constraints on theoretical models. I will then discuss the scientific and technological challenges that the CMB community must meet to ensure the success of future experiments.

Feb 22, Thursday
12 noon (BIDMaP Seminar)
Biwei Dai, UCB
373 Cory Hall and ZOOM
Deep Probabilistic Models for Cosmological Analysis and Beyond
Current and future weak lensing surveys contain significant information about our universe, but their optimal cosmological analysis is challenging, with traditional analyses often resulting in information loss due to reliance on summary statistics like two-point correlation functions. While deep learning methods offer promise in capturing the complex non-linear features of these cosmological fields, they often suffer from issues such as inadequate uncertainty quantification, susceptibility to distribution shifts, and interpretability limitations, which hinder their scientific applicability. In this talk, I propose a novel approach leveraging generative probabilistic modeling with Normalizing Flows to learn the data likelihood function at the field level, facilitating more effective cosmological information extraction. This framework not only enables anomaly detection of distribution shifts to improve the robustness of the analysis, but also fostering interpretability via generated samples. I will also discuss incorporating physical prior knowledge, such as symmetries and multiscale structure, into the model architectures to improve their generalization capabilities. Finally, I will explore the broader implications of deep probabilistic models in physics, highlighting their potential applications in diverse areas ranging from astronomical observations to high-energy physics and lattice field theory.

Feb 27, Tuesday
1:10 pm (BCCP/Cosmology seminar)
Eli Visbal, Univ of Toledo
Campbell 131
Revealing the First Billion Years of the Universe
Several key questions regarding the first billion years of the universe are currently unanswered: What were the properties of the first stars and galaxies? How did the first supermassive black holes form? How and when did reionization of the intergalactic gas occur? Fortunately, instruments such as the James Webb Space Telescope and line intensity mapping experiments can provide a wealth of data probing this epoch. In this talk, I will focus on some of the theoretical predictions required to interpret these observations. First, I will discuss our group’s work on hydrodynamical cosmological simulations of the first stars and semi-analytical models of the first galaxies. I will then discuss line-intensity mapping, which aims to measure cosmological large-scale structure through redshifted spectral lines. I will summarize recent radiative transfer simulations of Lyman-alpha intensity maps. These maps can be measured with instruments like SPHEREx to probe the cosmic reionization history. I will also describe how high-redshift intensity maps can be cross correlated with JWST data to ensure astrophysical contamination has been properly removed.

Feb 27, Tuesday
4pm LBL (RPM)
Brendan Crill, JPL
LBL 50A-5132 (Sessler room) and ZOOM
Sensitivity limits to cosmological constraints on early Universe physics: lessons from Planck for future measurements
The past several decades have seen enormous breakthroughs in the understanding of fundamental physics of the Universe through observations of the cosmic microwave background (CMB) and large scale structure. A synergy of rapid advances in technology, data analysis capabilities, and theory has led to an abundance of measurements, which in turn has seen the success of the Lambda-CDM model seeded by early-Universe inflation. This theory explains observations very few free parameters has emerged at explaining the observed structure of the universe.
Further refinements to this model and further characterization of observables in the CMB are a high priority for a deeper understanding of the nature of the physics of inflation and the evolution of structure across cosmic time. This will come about through measurements of non-Gaussianity, lensing, and the imprint of tensor modes in large-scale CMB polarization among other observables.
To achieve these refinements, projects have focused on improvements in raw sensitivity via new and more powerful detectors arrays with vastly more elements.
However, the success of earlier measurements has only come about through understanding and correction of systematic errors, understanding astrophysical foregrounds: more powerful instruments with better sensitivity are likely to uncover new challenges.
Here I describe lessons learned from the Planck mission at handling unforeseen instrumental and observational challenges and discuss prospects for fulfilling the sensitivity promise offered by large-scale structure projects such as SPHEREx and future CMB measurements.

Mar 5, Tuesday
1:10 pm (BCCP/Cosmology seminar)
Georgios Valogiannis, Chicago
Campbell 131
Precise Cosmological Constraints from BOSS Galaxy Clustering using the Wavelet Scattering Transform
Optimal extraction of the non-Gaussian information encoded in the Large-Scale Structure (LSS) of the universe lies at the forefront of modern precision cosmology. In this talk, I will discuss recent efforts to achieve this task using the Wavelet Scattering Transform (WST), which subjects an input field to a layer of non-linear transformations that are sensitive to non-Gaussianity through a generated set of WST coefficients. In order to assess its applicability in the context of LSS surveys, I will present the first WST application to actual galaxy observations, through a WST re-analysis of the BOSS DR12 CMASS dataset. After laying out the procedure on how to capture all necessary layers of realism for an application to data obtained from a spectroscopic survey, I will show results for the marginalized posterior probability distributions of multiple cosmological parameters obtained from a likelihood analysis of the CMASS data. A joint WST+ 2-point correlation function (2pcf) analysis is found to deliver a substantial improvement in the values of the predicted 1σ errors compared to the regular 2pcf-only analysis, highlighting the exciting prospect of employing higher-order statistics in order to fully exploit the potential of upcoming Stage-IV spectroscopic observations.

Mar 12, Tuesday
1:10 pm (BCCP/Cosmology seminar)
Sam Goldstein, Columbia
Campbell 131
Squeezing fNL out of the matter bispectrum with cosmological consistency relations
Constraining the amplitude of local primordial non-Gaussianity (fNL) is a key goal of upcoming large-scale structure (LSS) surveys. Constraints on fNL from LSS surveys typically take advantage of the distinct imprint of local primordial non-Gaussianity on the large-scale galaxy power spectrum via the “scale-dependent bias.” A complementary approach is to constrain fNL directly from the bispectrum; however, this requires a robust model for the gravitational contribution to the bispectrum. Analytically modeling this gravitational non-Gaussianity with perturbative techniques can be particularly challenging for local fNL, where the primordial bispectrum peaks in the squeezed limit, i.e., when two of the wavenumbers are much larger than the third. In this talk, I will present an analytic model for the non-linear squeezed matter bispectrum in the presence of local fNL based on the LSS consistency relations — non-perturbative statements about the structure of LSS correlation functions derived from symmetries of the LSS equations of motion. Using measurements from Nbody simulations with and without local primordial non-Gaussianity, I will demonstrate that our model can be used to infer unbiased constraints on fNL well beyond the range of scales that can be modeled using conventional techniques, such as perturbation theory. I will then show how our results can be used to derive an estimator for fNL from squeezed configurations of CMB lensing and cosmic shear bispectra. Finally, I will present and validate a non-perturbative model for the collapsed matter trispectrum in the presence of local fNL based on LSS consistency relations.

Mar 12, Tuesday
4pm LBL (RPM)
Colin Hill, Columbia
LBL 50A-5132 (Sessler room) and ZOOM
Uncovering Physics Beyond the Standard Model in the Cosmic Microwave Background`
The cosmic microwave background (CMB) anisotropy remains the cleanest, most powerful probe of fundamental physics in the cosmos. Measurements of the small-scale CMB temperature and polarization fields are set to undergo transformative improvements in the coming decade, culminating in the CMB-S4 experiment, which will open new windows into physics beyond the standard models (BSM) of particle physics and cosmology. I will first discuss my ongoing work in the Atacama Cosmology Telescope collaboration focused on cosmological parameter constraints from high-resolution CMB power spectra, with a particular emphasis on constraining BSM physics operating just prior to recombination, including new relativistic particles and new pseudo-scalar fields. I will describe high-precision end-to-end tests of our entire analysis pipeline on realistic, non-Gaussian sky simulations containing all relevant Galactic and extragalactic foregrounds, tightly coupled to a detailed instrument model. I will then turn to novel searches for BSM physics in CMB secondary anisotropies, as could be imprinted by the screening of CMB photons by massive dark photons or axion-like particles. I will show the first results of searches for these signals in CMB data, enabled by my group’s state-of-the-art needlet internal linear combination code. Finally, I will discuss new analysis methodologies for more optimally cleaning non-Gaussian foregrounds in CMB maps, with applications to both primordial B-mode searches and small-scale secondary anisotropy measurements. I will conclude with a look ahead to the prospects for BSM physics from the Simons Observatory and CMB-S4.

Mar 19, Tuesday
CANCELLED
Zack Li, UCB
Campbell 131

Mar 19, Tuesday
4pm LBL (RPM)
Christian Reichardt, Melbourne
LBL 50A-5132 (Sessler room) and ZOOM
Observing the oldest light in the Universe from the South Pole: Gravitational waves, Neutrinos and more!
The cosmic microwave background (CMB) radiation is the oldest light we can see. Since it bears the imprint of the universe just after the Big Bang (at a mere 0.003% of the universe’s age today), observations of the CMB are a crucial tool in our quest to understand how the Universe began and what its future holds. The South Pole is one of the best sites on Earth for these observations, and is the home of the South Pole Telescope (as well as CMB-S4 in the future). In this talk, I will present the latest CMB power spectra measurements from the SPT-3G instrument on the South Pole Telescope, and the resulting constraints on cosmology.

Mar 21, Thursday
3:40 pm (Astronomy Colloquium)
Adam Riess, JHU
Physics 1
JWST and the Hubble Tension

Mar 26, Tuesday
Spring Break, no 1 pm talk

Mar 26, Tuesday
4pm LBL (RPM)
Raphael Flauger, San Diego
LBL 50A-5132 (Sessler room) and ZOOM
Deciphering the Beginning
The cosmic microwave background (CMB) contains a wealth of information about cosmology as well as high-energy physics. It tells us about the composition and geometry of the universe, the properties of neutrinos, dark matter, and even the conditions in our universe long before the cosmic microwave background was emitted. After a brief introduction, I will discuss what we may hope to learn about the very early universe from upcoming CMB experiments that search for the imprint of primordial gravitational waves. One of the key challenges of this endeavor is the polarized emission from our galaxy. I will discuss our recent efforts to model and better understand the emission with the help of MHD simulations.

Apr 2, Tuesday
1:10 pm (BCCP/Cosmology seminar)
Tim Eifler, Arizona
Campbell 131
Constraining cosmology and baryonic physics with weak lensing, galaxy clustering, CMB lensing, and their cross-correlations
Baryonic feedback and cooling processes are a major modeling uncertainty for cosmological observables and interesting astrophysics in itself. For any multi-probe analysis including weak lensing, baryons are the main obstacle that prevents the inclusion of smaller/nonlinear scales that contain a wealth of information about structure formation and our cosmological model.
In this talk I will review constraints on baryonic physics (and cosmology) from a joint analysis of DES Year 1 weak lensing and galaxy clustering data in combination with CMB Lensing information from the Planck satellite mission (specifically https://arxiv.org/abs/2007.15026 , https://arxiv.org/abs/2311.08047 ). I will give an outlook to near-term extensions of this so-called 6x2 analysis based on DES Y3/Y6 and ACT CMB Lensing data. These analyses on existing data are natural precursors of future joint analysis using Rubin Observatory’s LSST and NASA’s Roman Space Telescope in combination with CMB Lensing from Simons Observatory and CMB-S4.

Apr 2, Tuesday
4pm LBL (RPM)
Boryana Hadzhiyska, UCB
LBL 50A-5132 (Sessler room) and ZOOM
Solving big-scale problems with small-scale models in the era of CMB-S4
In this talk, I will focus on some of the new opportunities and new challenges that CMB-S4, the next-generation cosmic microwave background (CMB) experiment, presents, and I will offer a viable path forward to meeting and exceeding the goals we have set up for it. The CMB has played a huge role in shaping our understanding of the history, evolution and contents of our Universe. From our observations of the CMB, we have attained a good grasp of about 13.8 billion years of cosmic history, but our direct and indirect observations are limited to a much smaller range in energy scale, which leaves unresolved some of the biggest puzzles such as the mechanism that seeded the primordial fluctuations and the nature of the dark sector. I will discuss how we can leverage observations of the CMB to learn about these much earlier periods of cosmic history and also, how the CMB provides complementary probes of the lower-redshift universe. On large scales, the B-mode polarization signal can reveal the origin of primordial fluctuations, whereas the small-scale signal holds information about the presence of non-standard light relics and the sum of the neutrino masses. However, measuring the primordial signal is a non-trivial task, as on its way to us, the signal has picked up contributions from lensing and astrophysical sources, known as foregrounds. The best way to design accurate approaches for mitigating these foregrounds is via realistic all-sky simulations. I will present new techniques for painting foregrounds onto large-volume simulations, using physically motivated models that close the loop between simulations and observations. In addition, I will discuss ways of optimizing the CMB lensing reconstruction techniques, which is crucial for improving our constraints on primordial quantities such as the number of light relics and the tensor-to-scalar ratio as well as the sum of neutrino masses. Finally, I will argue that synergies between CMB secondaries and large-scale structure surveys can place tight constraints on cosmology and astrophysics. The measurement precision of CMB-S4 will be unmatched, but in order to make the most out of it, we need a big step-up in our analysis and theory tools, including the development of realistic all-sky simulations as well as techniques for optimal extraction of the small-scale cosmological signal.

April 5, Friday
12 noon (INPA)
Eric Gawiser, Rutgers
LBL 50-5132 and ZOOM
Improving Photometric Redshifts for 3x2pt Cosmology: Training Sample Augmentation, Optimal Binning, and Neural Network Classifiers
Large imaging surveys of galaxies rely on photometric redshifts (photo-z’s) and tomographic binning for 3 × 2 pt analyses that combine galaxy clustering and weak lensing. We divide simulated galaxy catalogs into training and application sets, where the spectroscopic training set is non-representative in a realistic way, and then estimate photometric redshifts for the application set. Spectroscopic training samples for the Vera C. Rubin Observatory Legacy Survey of Space and Time (LSST) will be biased towards redder, brighter, lower-redshift galaxies, leading to photo-z estimates with outlier fractions nearly 4 times larger than for a representative training sample. Training sample augmentation allows us to add simulated galaxies possessing otherwise unrepresented features to our mock spectroscopic training sample, reducing the outlier fraction of the photo-z estimates by 50% and the scatter by 56%. We sort the galaxies into redshift bins chosen to maximize the 3x2pt signal using a novel generalized binning parameterization introduced by Moskowitz et al. (2023, ApJ 950, 49). Applying a neural network classifier trained to identify galaxies that are highly likely to be sorted into the correct redshift bin improves the figure of merit by ∼13%, equivalent to a 28% increase in data volume.

Apr 9, Tuesday
1:10 pm (BCCP/Cosmology seminar)
He Jia, Princeton
Campbell 131
Simulation-Based Inference of Large Scale Structure with Neural Quantile Estimation
While upcoming cosmological surveys promise to map the universe with unprecedented precision, the challenge remains on how to optimally extract information from this wealth of data. We introduce Neural Quantile Estimation (NQE), a novel Simulation-Based Inference method based on conditional quantile regression, and its application to the field level inference of cosmological large scale structure. NQE autoregressively learns one-dimensional quantiles for each posterior dimension, conditioned on the observation data and previous posterior dimensions. When provided with sufficient training data, NQE converges to the Bayesian optimal posterior. In scenarios with limited training data, a post-processing step can be employed to ensure the posterior remains well-calibrated, with minimal computational overhead. Moreover, such post-processing calibration can mitigate biases stemming from model misspecification, notably those associated with inaccurate surrogate models and/or baryonic physics uncertainties.

Apr 9, Tuesday
4pm LBL (RPM)
Michel Piat, APC-IN2P3
LBL 50A-5132 (Sessler room) and ZOOM
The Q&U Bolometric Interferometer for Cosmology (QUBIC): status and perspectives
QUBIC (Q & U Bolometric Interferometer for Cosmology) is an international ground-based experiment dedicated in the measurement of the polarized fluctuations of the millimeter sky and especially the Cosmic Microwave Background (CMB). It is based on bolometric interferometry, an original detection technique which combine the immunity to systematic effects of an interferometer with the sensitivity of low temperature incoherent detectors. QUBIC is currently in commissioning in Argentina, at the Alto Chorrillos mountain site at 5000m a.s.l. near San Antonio de los Cobres, in the Salta province. After a description of the QUBIC instrument, I will present the last results of the first characterization phases with a focus on the detectors and readout system based on NbSi Transition Edge Sensors (TESs) cooled to 320mK. The perspective of spectro-imaging will also be described allowing for better foreground mitigation. This feature will soon be tested with QUBIC and could represent a significant breakthrough in CMB polarimetry.

April 12, Friday
12 noon (INPA)
Elisabeth Krause, Arizona
LBL 50-5132 and ZOOM
A Parameter-Masked Mock Data Challenge for Beyond-Two-Point Galaxy Clustering Statistics
The last few years has seen the emergence of a vast array of novel techniques for the analysis of high precision data from upcoming galaxy surveys, which are primarily motivated by the notion that any optimal analysis of galaxy clustering data should extend beyond the canonical two-point (2pt) statistics. We test and benchmark some of these new techniques in a community data challenge ``Beyond-2pt’’. The challenge dataset consists of high-precision mock galaxy catalogs for clustering in real-space, redshift-space, and on a light cone. Participants in the challenge have developed end-to-end pipelines to analyze mock catalogs and extract unknown (``masked'') cosmological parameters of the underlying LCDM models with their methods. The methods represented are density-split clustering, nearest neighbor statistics, BACCO power spectrum emulator, void statistics, field-level effective field theory (EFT), and joint power spectrum and bispectrum analyses using both EFT and simulation-based inference.
I will review the results of the challenge, focusing on problems solved, lessons learned, and future research needed to perfect the emerging beyond-2pt approaches.

Apr 16, Tuesday
1:10 pm (BCCP/Cosmology seminar)
Nivedita Mahesh, Caltech
Campbell 131
21cm cosmology: Efforts with EDGES, OVRO-LWA and DSA-2000
Standard cosmological models predict that the first astrophysical sources formed from a Universe filled with neutral hydrogen (HI) around one hundred million years after the Big Bang. The transition into Cosmic Dawn (CD) that seeded all the structures we see today can only be probed directly by the 21-cm line of HI. Redshifted by the Hubble expansion, HI emission, and absorption during CD is expected to be visible in radio frequencies between 40 and 100 MHz, tracing redshifts of 35 >z>13. Precisely characterized and carefully calibrated low-frequency instruments are necessary to measure the predicted ~ 10-200 mK brightness temperature of this redshifted 21-cm signal against bright galactic foregrounds. In this talk, I will discuss two experiments I work with, designed to probe the Cosmic Dawn - EDGES and OVRO-LWA. I will touch on the efforts carried out by the EDGES team since the reported detection in Bowman et al. (2018), focusing on new analysis techniques and updates on the new EDGES 3 system. For the OVRO- Long Wavelength Array, I will present the details of the upgraded telescope and discuss the development efforts of the 21cm cosmology pipeline. I will briefly shift gears to "lower" redshifts, highlighting the potential of Large N small D arrays (like DSA 2000) to perform wide and deep 21cm intensity mapping surveys that can be used to probe late-time cosmology.

Apr 16, Tuesday
4pm LBL (RPM)
Julien Guy, LBL
LBL 50A-5132 (Sessler room) and ZOOM
The Dark Energy Spectroscopic Instrument First Year Results: Cosmic Expansion History with Baryon Acoustic Oscillations
The Dark Energy Spectroscopic Instrument (DESI) collaboration is conducting a 5 year redshift survey of 40 million extra-galactic sources over 14,000 square degrees of the northern sky. One of its primary goals is to measure the cosmic expansion history with baryon acoustic oscillations (BAO). I will present the measurement of BAO in galaxy, quasar and Lyman-alpha forest tracers from the first year of observation. With 5.7 million galaxy and quasar redshifts in the range 0.1 < z < 2.1, and 420,000 Lyman-alpha forest quasars at higher redshift, the aggregate precision on BAO is of 0.52% at z<2.1 and 1.1% at an effective redshift z=2.3, surpassing in a year two decades of observations with the SDSS. I will present some of the numerous validation tests performed with simulations and blinded data. I will then highlight the main cosmological results, with improved constraints on the dark energy equation of state, the Hubble parameter, spatial curvature, and the sum of neutrino masses.

Apr 23, Tuesday
1:10 pm (BCCP/Cosmology seminar)
Kendrick Smith, Perimeter
Campbell 131
FRB science results from CHIME
Fast radio bursts (FRBs) are a recently discovered, poorly understood class of transient event, and understanding their origin has become a central problem in astrophysics. I will present FRB science results from CHIME, a new interferometric telescope at radio frequencies 400-800 MHz. Since 2018, CHIME has found ~20 times more FRBs than all other telescopes combined, including ~60 new repeating FRBs, the first repeating FRB with periodic activity, an FRB pulse in our own galaxy from a known magnetar, and millisecond periodicity in FRB sub-pulses. These results were made possible by new algorithms which can be used to build radio telescopes orders of magnitude more powerful than CHIME. I will briefly describe two upcoming projects: outrigger telescopes for CHIME (starting this year) and CHORD, a new telescope with ~10 times the CHIME mapping speed (starting in 2026).

May 3, Friday
12 noon (INPA)
Caroline Cuesta Lazaro, MIT
LBL 50-5132 and ZOOM
Big Data cosmology meets AI
The upcoming era of cosmological surveys promises an unprecedented wealth of observational data that will transform our understanding of the universe. Surveys such as DESI, Euclid, and the Vera C. Rubin Observatory will provide extremely detailed maps of billions of galaxies out to high redshifts. Analyzing these massive datasets poses exciting challenges that machine learning is uniquely poised to help overcome. In this talk, I will highlight recent examples from my work on probabilistic machine learning for cosmology. First, I will explain how a point cloud diffusion model can be used both as a generative model for 3D maps of galaxy clustering and as a likelihood model for such datasets. Moreover, I will present a generative model developed to reconstruct the dark matter cosmic web from biased galaxy clustering observations, in a probabilistic manner. And finally, I will introduce ongoing work on developing fast, differentiable, and accurate hybrid physics-ML simulators for N-body and hydrodynamical simulations. When combined with the wealth of data from upcoming surveys, these machine learning techniques have the potential to provide new insights into fundamental questions about the nature of the universe.

June 11, Tuesday
11:00 am
Parth Nayak, LMU Munich
LBL 50F-1647 and ZOOM
LyαNNA: A Deep Learning Field-level Inference Machine for the Lyman-α Forest
The Lyα forest, a congregation of absorption lines in the observed spectra of distant quasars imprinted by the intervening intergalactic medium (IGM), is a powerful probe of cosmic structures and thermodynamic properties of the universe at cosmological redshifts of z ~ 2 - 6. The inference of those characteristics from the Lyα forest conventionally relies on established summary statistics of the transmitted flux field, however, this inevitably leads to loss of some relevant parts of the information carried by the field. I will talk about a deep learning approach we have developed (described in this paper: https://arxiv.org/abs/2311.02167) in order to help remedy this problem for large future datasets. This framework consists of a 1D ResNet that extracts all the pertinent features directly from the field and optimally compresses them into statistical vectors, thereby facilitating Bayesian inference. We observe factors-of-a-few improvement in the posterior constraints on the IGM thermal parameters over the most commonly-used summary statistics and their combinations, when employing our machinery. I will also briefly touch upon the ongoing efforts to include observational and modelling systematic uncertainties in the data for creating a more robust pipeline. Our case study demonstrates the untapped potential of deep learning techniques for inference from large and complex datasets at field-level in astrophysics and cosmology.