ML | Raghav Kansal

Higgs Searches

Mon, 01 Jun 2020 00:00:00 +0000

I’m interested in precision measurements of the Higgs boson, as well as searches for new Higgs-like particles to explain mysteries such as baryon asymmetry. I’m currently leading the analysis of 2016-2018 CMS data looking for two high-momentum Higgs bosons (H) decaying to beauty quarks (b) and vector bosons (V). For this, in collaboration with Peking University and Fermilab, I developed a state-of-the-art transformer network to classify between H→VV particle clouds and backgrounds.

I’m leading as well the search for new Higgs-like particles (X, Y) decaying to beauty quarks and vector bosons. We expect to set strong limits on both Higgs to VV couplings and X→HY cross sections.

I have also been involved in di-Higgs searches in the 4b final state, the most sensitive measurement to date to di-Higgs production and the two-vector-boson coupling.

ML for Fast Simulations

Sat, 08 Jun 2019 00:00:00 +0000

Developed a graph-based generative adversarial network, MPGAN, which proved to be state-of-the-art at simulating particle collisions. Developed as well the faster attention-based generative adversarial particle transformer (GAPT), using set transformers, and efficient and sensitive two-sample goodness-of-fit tests for validating fast simulations. Working now on extending to conditional generation and application to detector data.

Equivariant Neural Networks

Sun, 27 Dec 2020 00:00:00 +0000

Developed a graph-based autoencoder (PGAE) for compression of and anomaly detection in Large Hadron Collider data. Wrote a review of deep learning models that are equivariant to physics-relevant group transformations for the UCSD group theory course. Led to developing a graph-based autoencoder equivariant to Lorentz group transformations as well (LGAE). LGAE outperformed CNN and GNN autoencoders for compression and anomaly detection tasks.

JetNet

Wed, 01 Sep 2021 00:00:00 +0000

Developed a library for convenient access to jet datasets, and other utilities, to increase accessibility and reproducibility in ML in particle physics. >35,000 downloads as of September 2023, used in several ML and particle physics projects.

Anomaly Detection

Wed, 01 Dec 2021 00:00:00 +0000

Developed several approaches for finding rare particle collisions, including GNNs, Lorentz-equivariant networks, and multi-variate goodness-of-fit tests. More info coming soon!

Explainable AI

Sat, 01 May 2021 00:00:00 +0000

Interpreting results of machine learning models for reconstruction and jet classification using explainable AI techniques. More info coming soon!

Sequential Modeling for Soccer Predictions

Sat, 27 Apr 2019 00:00:00 +0000

JetNet: A Python package for accessing open datasets and benchmarking machine learning methods in high energy physics

Fri, 01 Sep 2023 00:00:00 +0000

Fast jet simulations and how to evaluate them

Tue, 01 Aug 2023 08:00:00 +0000

Boosted multi-Higgs with jets measurements in CMS

Mon, 31 Jul 2023 08:00:00 +0000

Machine learning for particle physics simulations

Mon, 24 Jul 2023 08:00:00 +0000

Generative transformers and how to evaluate them (+ Lorentz-equivariant networks)

Tue, 27 Jun 2023 08:00:00 +0000

Lorentz group equivariant autoencoders

Fri, 09 Jun 2023 00:00:00 +0000

Applications of two-sample goodness-of-fit tests to deep generative models

Thu, 01 Jun 2023 08:00:00 +0000

Evaluation Metrics for FastSim

Wed, 31 May 2023 08:00:00 +0000

Multivariate goodness of fit testing for evaluating HEP generative models

Wed, 31 May 2023 08:00:00 +0000

Multivariate goodness of fit testing for evaluating HEP generative models

Mon, 22 May 2023 08:00:00 +0000

Evaluating generative models in high energy physics

Sat, 01 Apr 2023 00:00:00 +0000

Generative transformers and how to evaluate them

Tue, 14 Feb 2023 08:00:00 +0000

FAIR AI Models in High Energy Physics

Wed, 21 Dec 2022 00:00:00 +0000

FastSim on GPUs

Tue, 13 Dec 2022 08:00:00 +0000

Do graph neural networks learn traditional jet substructure?

Fri, 09 Dec 2022 00:00:00 +0000

Generative transformers and how to evaluate them

Mon, 21 Nov 2022 08:00:00 +0000

On the Evaluation of Generative Models in HEP

Tue, 01 Nov 2022 08:00:00 +0000

JetNet library for machine learning in high energy physics

Fri, 16 Sep 2022 08:00:00 +0000

Discussion on Generative Models

Wed, 14 Sep 2022 08:00:00 +0000

Particle Cloud Generation with Message Passing GANs

Fri, 09 Sep 2022 08:00:00 +0000

Overview and Outlook: Machine Learning for Simulation

Thu, 21 Jul 2022 15:00:00 +0000

Particle Cloud Generation with Message Passing GANs

Sun, 17 Jul 2022 08:00:00 +0000

Machine Learning for LHC Simulations

Thu, 14 Jul 2022 16:00:00 +0000

Particle-based fast jet simulation at the LHC with variational autoencoders

Wed, 13 Jul 2022 00:00:00 +0000

Search for nonresonant pair production of highly energetic Higgs bosons decaying to bottom quarks

Wed, 01 Jun 2022 00:00:00 +0000

Particle Cloud Generation with Message Passing GANs

Sun, 10 Apr 2022 08:00:00 +0000

Search for boosted Higgs boson pair production in the bbVV all-hadronic final state in CMS

Sat, 09 Apr 2022 08:00:00 +0000

Improving Di-Higgs Sensitivity at Future Colliders in Hadronic Final States with Machine Learning

Mon, 04 Apr 2022 00:00:00 +0000

A FAIR and AI-ready Higgs boson decay dataset

Mon, 14 Feb 2022 00:00:00 +0000

Particle Cloud Generation with Message Passing GANs

Fri, 03 Dec 2021 08:00:00 +0000

Particle Cloud Generation with Message Passing Generative Adversarial Networks

Fri, 03 Dec 2021 00:00:00 +0000

Particle Graph Autoencoders and Differentiable, Learned Energy Mover's Distance

Thu, 02 Dec 2021 00:00:00 +0000

Explaining machine-learned particle-flow reconstruction

Wed, 01 Dec 2021 00:00:00 +0000

Particle Cloud Generation with Message Passing GANs

Sun, 28 Nov 2021 00:00:00 +0000

Validation Techniques for Machine-Learned FastSim

Tue, 23 Nov 2021 08:00:00 +0000

Particle Cloud Generation with Message Passing GANs

Mon, 08 Nov 2021 08:00:00 +0000

Particle Cloud Generation with Message Passing GANs

Mon, 11 Oct 2021 08:00:00 +0000

Sparse Data Generation for Particle-Based Simulation of Hadronic Jets in the LHC

Tue, 21 Sep 2021 00:00:00 +0000

Particle Cloud Generation with Message Passing GANs

Wed, 07 Jul 2021 08:00:00 +0000

Particle Cloud Generation with Message Passing GANs

Wed, 23 Jun 2021 08:00:00 +0000

Sparse Data Generation

Wed, 05 May 2021 07:00:00 +0000

Graph GANs for High Energy Physics Data Generation

Thu, 18 Mar 2021 13:00:00 +0000

Graph GANs for High Energy Physics Data Generation

Wed, 17 Mar 2021 13:00:00 +0000

Graph GANs for High Energy Physics Data Generation

Tue, 02 Feb 2021 08:00:00 +0000

Graph GANs for Sparse Data Generation in High Energy Physics

Fri, 11 Dec 2020 13:00:00 +0000

Graph Generative Adversarial Networks for Sparse Data Generation in High Energy Physics

Mon, 30 Nov 2020 00:00:00 +0000

Sparse Data Generation with Graph GANs

Mon, 23 Nov 2020 08:20:00 +0000

Deep Graph Neural Networks for Fast HGCAL Simulation

Thu, 27 Feb 2020 17:00:00 +0000

Deep Graph Neural Networks for Fast HGCAL Simulation

Wed, 28 Aug 2019 16:15:00 +0000

Deep Graph Neural Networks for Fast HGCAL Simulation

Wed, 21 Aug 2019 13:40:00 +0000

Deep Graph Neural Networks for Fast HGCAL Simulation

Tue, 13 Aug 2019 13:38:00 +0000