Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

AGUFM

Comparisons of a Quantum Annealing and Classical Computer Neural Net Approach for Inferring Global Annual CO2 Fluxes over Land

Milton Halem, A. Radov, David Joseph Singh

Published Dec 1, 2017• 0 citations
OpenAlex

Related papers

A Quantum Annealing Computer Team Addresses Climate Change Predictability

Milton Halem, J. LeMoigne, John E. Dorband +13 more

The near confluence of the successful launch of the Orbiting Carbon Observatory2 on July 2, 2014 and the acceptance on August 20, 2015 by Google, NASA Ames Research Center and USRA of a 1152 qubit D-Wave 2X Quantum Annealing Computer (QAC), offered an exceptional opportunity to explore the potential of this technology to address the scientific prediction of global annual carbon uptake by land surface processes. At UMBC,we have collected and processed 20 months of global Level 2 light CO₂ data as well as fluorescence data. In addition we have collected ARM data at 2sites in the US and Ameriflux data at more than 20 stations. J. Dorband has developed and implemented a multi-hidden layer Boltzmann Machine (BM) algorithm on the QAC. Employing the BM, we are calculating CO₂ fluxes by training collocated OCO-2 level 2 CO₂ data with ARM ground station tower data to infer to infer measured CO₂ flux data. We generate CO₂ fluxes with a regression analysis using these BM derived weights on the level 2 CO₂ data for three Ameriflux sites distinct from the ARM stations. P. Gentine has negotiated for the access of K34 Ameriflux data in the Amazon and is applying a neural net to infer the CO₂ fluxes. N. Talik validated the accuracy of the BM performance on the QAC against a restricted BM implementation on the IBM Softlayer Cloud with the Nvidia co-processors utilizing the same data sets. G. Nearing and K. Harrison have extended the GSFC LIS model with the NCAR Noah photosynthetic parameterization and have run a 10 year global prediction of the net ecosystem exchange. C. Pellisier is preparing a BM implementation of the Kalman filter data assimilation of CO₂ fluxes. At UMBC, R. Prouty is conducting OSSE experiments with the LISNoah model on the IBM iDataPlex to simulate the impact of CO₂ fluxes to improve the prediction of global annual carbon uptake. J. LeMoigne and D. Simpson have developed a neural net image registration system that will be used for MODIS ENVI and will be converted to a BM algorithm implementation on the QAC. The first integer adder has been implemented on the D-Wave 2X by A. Shehab that will perform HAAR wavelets for image compression of MODIS scenes. Finally, based on the next generations of QACs, we are preparing a 5-year performance road map on the scalability of the current QAC algorithms.

•Jun 14, 2016•0 citations

Leveraging Quantum Computing for Climate Change Modeling: A Path to Accurate Global Predictions

Suhaib Rehman, Rakesh Thakur, Raghuvaran Kendyala +4 more

Lecture notes in networks and systems•Jan 1, 2026•0 citations

How Quantum Computing-Friendly Multispectral Data can be?

Manish K. Gupta, Martin Beseda, Piotr Gawron

Quantum computers consisting of hundreds of noisy qubits are already available and can run specific quantum algorithms although a large-scale fully error-corrected quantum computer is decades away. It is important to study their application to real-life computational problems. One such problem is Land Use and Land Cover classification of Earth Observation data set collected from the earth observation satellite mission using quantum machine learning methods. In this work, we compare the performance of the classical neural network on the re-labeled dataset of the Copernicus Sentinel-2 mission, when the model has access to Projected Quantum Kernel features. We show that classical neural net-work training accuracy increases drastically when the model has access to Projected Quantum Kernel features. This study shows the potential for quantum machine learning methods to Earth Observation data and provides key evidence for further investigation.

IEEE International Geoscience and Remote Sensing Symposium (IGARSS)•Jul 17, 2022•23 citations

Quantum annealing for remote sensing data processing: A review of optimization based applications

Indranil Misra, S. Manthira Moorthi, Debajyoti Dhar

Remote Sensing Applications Society and Environment•Jan 1, 2026•3 citations

Completely quantum neural networks

Steven Adam Abel, Juan Carlos Criado, Michael Spannowsky

Artificial neural networks are at the heart of modern deep learning algorithms. We describe how to embed and train a general neural network in a quantum annealer without introducing any classical element in training. To implement the network on a state-of-the-art quantum annealer, we develop three crucial ingredients: binary encoding the free parameters of the network; polynomial approximation of the activation function; and reduction of binary higher-order polynomials into quadratic ones. Together, these ideas allow encoding the loss function as an Ising model Hamiltonian. The quantum annealer then trains the network by finding the ground state. We implement this for an elementary network and illustrate the advantages of quantum training: its consistency in finding the global minimum of the loss function and the fact that the network training converges in a single annealing step, which leads to short training times while maintaining a high classification performance. After training the network using a quantum annealer, one can then use the quantum network weights in a classical network algorithm of identical design for inference. Our approach opens an avenue for the quantum training of general machine learning models.

arXiv:2202.11727•Physical Review A•Aug 1, 2022•46 citations

Data Assimilation on a Quantum Annealing Computer: Feasibility and Scalability

Grey Nearing, Milton Halem, D. R. Chapman +1 more

2014 AGU Fall Meeting•Dec 15, 2014•1 citation
The near confluence of the successful launch of the Orbiting Carbon Observatory2 on July 2, 2014 and the acceptance on August 20, 2015 by Google, NASA Ames Research Center and USRA of a 1152 qubit D-Wave 2X Quantum Annealing Computer (QAC), offered an exceptional opportunity to explore the potential of this technology to address the scientific prediction of global annual carbon uptake by land surface processes. At UMBC,we have collected and processed 20 months of global Level 2 light CO₂ data as well as fluorescence data. In addition we have collected ARM data at 2sites in the US and Ameriflux data at more than 20 stations. J. Dorband has developed and implemented a multi-hidden layer Boltzmann Machine (BM) algorithm on the QAC. Employing the BM, we are calculating CO₂ fluxes by training collocated OCO-2 level 2 CO₂ data with ARM ground station tower data to infer to infer measured CO₂ flux data. We generate CO₂ fluxes with a regression analysis using these BM derived weights on the level 2 CO₂ data for three Ameriflux sites distinct from the ARM stations. P. Gentine has negotiated for the access of K34 Ameriflux data in the Amazon and is applying a neural net to infer the CO₂ fluxes. N. Talik validated the accuracy of the BM performance on the QAC against a restricted BM implementation on the IBM Softlayer Cloud with the Nvidia co-processors utilizing the same data sets. G. Nearing and K. Harrison have extended the GSFC LIS model with the NCAR Noah photosynthetic parameterization and have run a 10 year global prediction of the net ecosystem exchange. C. Pellisier is preparing a BM implementation of the Kalman filter data assimilation of CO₂ fluxes. At UMBC, R. Prouty is conducting OSSE experiments with the LISNoah model on the IBM iDataPlex to simulate the impact of CO₂ fluxes to improve the prediction of global annual carbon uptake. J. LeMoigne and D. Simpson have developed a neural net image registration system that will be used for MODIS ENVI and will be converted to a BM algorithm implementation on the QAC. The first integer adder has been implemented on the D-Wave 2X by A. Shehab that will perform HAAR wavelets for image compression of MODIS scenes. Finally, based on the next generations of QACs, we are preparing a 5-year performance road map on the scalability of the current QAC algorithms.
Quantum computers consisting of hundreds of noisy qubits are already available and can run specific quantum algorithms although a large-scale fully error-corrected quantum computer is decades away. It is important to study their application to real-life computational problems. One such problem is Land Use and Land Cover classification of Earth Observation data set collected from the earth observation satellite mission using quantum machine learning methods. In this work, we compare the performance of the classical neural network on the re-labeled dataset of the Copernicus Sentinel-2 mission, when the model has access to Projected Quantum Kernel features. We show that classical neural net-work training accuracy increases drastically when the model has access to Projected Quantum Kernel features. This study shows the potential for quantum machine learning methods to Earth Observation data and provides key evidence for further investigation.
Artificial neural networks are at the heart of modern deep learning algorithms. We describe how to embed and train a general neural network in a quantum annealer without introducing any classical element in training. To implement the network on a state-of-the-art quantum annealer, we develop three crucial ingredients: binary encoding the free parameters of the network; polynomial approximation of the activation function; and reduction of binary higher-order polynomials into quadratic ones. Together, these ideas allow encoding the loss function as an Ising model Hamiltonian. The quantum annealer then trains the network by finding the ground state. We implement this for an elementary network and illustrate the advantages of quantum training: its consistency in finding the global minimum of the loss function and the fact that the network training converges in a single annealing step, which leads to short training times while maintaining a high classification performance. After training the network using a quantum annealer, one can then use the quantum network weights in a classical network algorithm of identical design for inference. Our approach opens an avenue for the quantum training of general machine learning models.