skip to content
Home

Cavendish Astrophysics

 

Feed aggregator Subscribe to Cavendish Astrophysics aggregator

Wed 08 Oct 13:15: High-Dimension Bayesian Model Comparison in Cosmology

Upcoming talks - Mon, 06/10/2025 - 12:49
High-Dimension Bayesian Model Comparison in Cosmology

Our recent work (2509.13307) demonstrated the performance of GPU -accelerated nested sampling for efficient high-dimensional Bayesian inference in cosmology. Using JAX -based emulators and likelihoods we can leverage the parallel computing of GPUs to achieve orders of magnitude speed-ups against CPU -based analyses, and bring robust evidence calculations up to GPU -speed. This puts nested sampling back on equal footing with Markov Chain Monte Carlo (MCMC) methods, which use auto-diff gradients to achieve their speed-ups. In particular a Euclid-like mock Cosmic Shear likelihood has been considered, an analysis which previously took 8 months on a CPU instance, and we bring the time to constrain both ΛCDM and w0wa down to only 2 days on a single GPU . This talk explores a few options for pushing our methodology even further, in preparation for joint analyses of next generation of cosmlogical surveys.

Add to your calendar or Include in your list

Wed 08 Oct 13:40: ECHO21: A tool for modelling global 21-cm signal from dark ages to reionization

Upcoming talks - Mon, 06/10/2025 - 10:14
ECHO21: A tool for modelling global 21-cm signal from dark ages to reionization

I will introduce a Python package called ECHO21 for modelling the global 21-cm signal from the dark ages through cosmic dawn to the end of reionization. Leveraging its analytical framework, ECHO21 generates a single model in O(1) s, allowing a large number of signals to be generated efficiently by distributing models across multiple cores. Thus, it is ideal for performing astrophysical or cosmological inference from a given 21-cm dataset. We offer six astrophysical parameters that control the Lyman-α emissivity, X-ray emissivity, emissivity of ionizing photons, and star formation rate. Beyond its efficiency, some of the attractive and novel features in ECHO21 relative to previously published codes are the inclusion of Lyα heating, the ability to vary the standard cosmological parameters as easily as the astrophysical parameters, different models of star formation rate density (physically-motivated, a semi-empirical, and an empirically-motivated), and modelling the global signal for a Coulomb-like interacting DM (IDM) framework. This IDM model incorporates cooling of baryons as well as a delay in star formation. With several 21-cm experiments soon to provide cosmic dawn 21-cm data, ECHO21 is a flexible and extensible new open-source package for making quick and sufficiently realistic astrophysical inferences.

Add to your calendar or Include in your list

Wed 08 Oct 13:15: High-Dimension Bayesian Model Comparison in Cosmology

Upcoming talks - Mon, 06/10/2025 - 10:12
High-Dimension Bayesian Model Comparison in Cosmology

Abstract not available

Add to your calendar or Include in your list

Deputy Health and Safety Manager

Department of Physics Jobs - Mon, 06/10/2025 - 01:00

The Department of Physics (Cavendish Laboratory) is one of the best-performing departments at the University and in the UK. The Cavendish Laboratory conducts cutting edge physics research and high-level teaching of undergraduate and postgraduate students. We achieve excellence by working to provide an inclusive, safe, and healthy work environment as well as recruiting the best people to carry out and support teaching and research.

We are looking for an experienced and enthusiastic safety practitioner to join our small team as deputy to the Departmental Safety Manager (DSM). The role holder will assist the DSM in supporting academics, researchers, managers, staff, students, and others to understand and fulfil their health and safety responsibilities. The role holder will also report on health and safety to the Head of Department.

Under the leadership of the DSM, you will be providing guidance on safety management and risk control, supported by people with expertise in managing risks from biological materials, chemicals, and radiation. As the Department is large with a variety of significant risks, we are seeking someone with demonstrable experience of managing health and safety in a multi-site, multi-disciplinary environment, with a good track record of improving safety systems and culture. There is scope for development and career progression in the department for an enthusiastic, hardworking individual, and training will be provided.

You will be educated preferably to degree level or equivalent in a scientific discipline. You will also have demonstrable knowledge of health and safety management systems and standards, a proven track record of dealing with all aspects of H&S, experience of providing training and coaching, and an understanding of how to successfully merge good safety practice with good research and teaching. Experience of working in a scientific research environment would be an advantage but is not essential.

You will be able to communicate well with all levels of staff and students, deal with changing demands, solve problems, and organize your work priorities based on risks. You will be keen to learn, self-motivated and able to work well independently and as part of a small team, taking the lead in risk control where necessary.

The successful applicant will be required to complete training provided by the University Safety Office and become fully familiar with the significant risk issues in the Department.

Click the 'Apply' button below to register an account with our recruitment system (if you have not already) and apply online.

Informal enquiries are welcomed and should be directed to Geoff Elliott at gde26@phy.cam.ac.uk

Please quote reference KA47543 on your application and in any correspondence about this vacancy.

The University actively supports equality, diversity and inclusion and encourages applications from all sections of society.

The University has a responsibility to ensure that all employees are eligible to live and work in the UK.

Wed 05 Nov 14:00: Title to be confirmed

Upcoming talks - Fri, 03/10/2025 - 16:43
Title to be confirmed

Abstract not available

Add to your calendar or Include in your list

Fri 28 Nov 13:00: (TBC)

Upcoming talks - Fri, 03/10/2025 - 15:58
(TBC)

(TBC)

Add to your calendar or Include in your list

Fri 28 Nov 13:00: (TBC)

Upcoming talks - Fri, 03/10/2025 - 15:58
(TBC)

(TBC)

Add to your calendar or Include in your list

Fri 21 Nov 13:00: (TBC)

Upcoming talks - Fri, 03/10/2025 - 15:57
(TBC)

(TBC)

Add to your calendar or Include in your list

Fri 14 Nov 13:00: (TBC)

Upcoming talks - Fri, 03/10/2025 - 15:56
(TBC)

(TBC)

Add to your calendar or Include in your list

Fri 14 Nov 13:00: (TBC)

Upcoming talks - Fri, 03/10/2025 - 15:56
(TBC)

(TBC)

Add to your calendar or Include in your list

Fri 31 Oct 13:00: (TBC)

Upcoming talks - Fri, 03/10/2025 - 15:55
(TBC)

(TBC)

Add to your calendar or Include in your list

Fri 07 Nov 13:00: (TBC)

Upcoming talks - Fri, 03/10/2025 - 15:55
(TBC)

(TBC)

Add to your calendar or Include in your list

Fri 24 Oct 13:00: (TBC)

Upcoming talks - Fri, 03/10/2025 - 15:52
(TBC)

(TBC)

Add to your calendar or Include in your list

Fri 17 Oct 13:00: (TBC)

Upcoming talks - Fri, 03/10/2025 - 15:50
(TBC)

(TBC)

Add to your calendar or Include in your list

Fri 10 Oct 13:00: Advancing black-hole perturbation theory beyond the linear regime with the hyperboloidal framework

Upcoming talks - Fri, 03/10/2025 - 15:47
Advancing black-hole perturbation theory beyond the linear regime with the hyperboloidal framework

The hyperboloidal framework builds upon Penrose’s seminal work on the “Conformal Treatment of Infinity.” It’s an elegant concept that enables us to approach the black hole event horizon and the wave zone “at the same time”. Over the decades, it has become an indispensable tool in black hole perturbation theory, particularly when considering an expansion beyond the linear regime. In this presentation, I will delve into the latest advancements in the field. This includes introducing the so-called Minimal Gauge, a simple strategy to construct hyperboloidal slices and exploring its applications in gravitational wave physics. After reviewing contributions to fundamental concepts in the ringdown phase, I’ll discuss technical advances in the modeling of extreme mass ratio inspirals through the self-force program.

Add to your calendar or Include in your list

Thu 09 Oct 12:00: Einstein metrics, Interacting QFT’s and Confinement in four and Five dimensions

Upcoming talks - Fri, 03/10/2025 - 15:45
Einstein metrics, Interacting QFT’s and Confinement in four and Five dimensions

M-theory provides a geometric framework to describe a variety of interesting quantum field theories in which the QFT ’s arise from Einstein metrics. We motivate a precise definition of this framework which (partly) takes the form of the space of complete, asymptotically conical Ricci flat manifolds in various dimensions. We show how this provides insights into various strongly coupled systems such as non-Abelian gauge theories in four and more dimensions and leads to confining string theories in four and five dimensions. The four dimensional strings can be compared to flux tubes in Yang-Mills theories.

Add to your calendar or Include in your list

Tue 28 Oct 16:00: Title to be confirmed

Upcoming talks - Fri, 03/10/2025 - 15:12
Title to be confirmed

Abstract not available

Add to your calendar or Include in your list

Wed 15 Oct 14:00: TBD

Upcoming talks - Fri, 03/10/2025 - 15:10
TBD

Abstract not available

Add to your calendar or Include in your list

Mon 27 Oct 16:00: TBD

Upcoming talks - Fri, 03/10/2025 - 11:47
TBD

TBD

Add to your calendar or Include in your list

Tue 21 Oct 13:00: The architecture of planetary systems: from physical models to AI

None - Fri, 03/10/2025 - 11:35
The architecture of planetary systems: from physical models to AI

Present-day astronomical facilities allow detecting and characterizing Earth twins, provided we know where to search. In this context, understanding the correlations between properties of planets in the same system, also called the architecture of planetary systems, can provide valuable guidance to optimize observational campaigns. In this talk, I will present models of planetary system formation, the so-called ‘Bern model’, and how they can be used to train generative AI models. I will conclude by presenting an example of predictions made by these new physically informed AI models.

Add to your calendar or Include in your list

Contact us

Battcock Centre for Experimental Astrophysics
Cavendish Laboratory
JJ Thomson Avenue
Cambridge CB3 0HE

Site Privacy & Cookie Policies

Quick links

Study at Cambridge

About the University

Research at Cambridge