Atreya Majumdar

Atreya Majumdar
Position | Junior Research Leader |
atreya.majumdar@uni-due.de | |
Phone | 0203 3379 4725 |
Office | MG 392 |
Address | Twist Group, Faculty of Physics University of Duisburg-Essen Campus Duisburg Lotharstraße 1 D 47057 Duisburg |
Contact | ORCID-ID: 0000-0002-6547-2231 Google Scholar Profile |
Fakultät für Physik, Theoretische Physik
47057 Duisburg
Functions
-
Wissenschaftliche/r Mitarbeiter/in, Arbeitsgruppe Prof. Everschor-Sitte
Current lectures
-
SoSe 2025
Past lectures (max. 10)
The following publications are listed in the online university bibliography of the University of Duisburg-Essen. Further information may also be found on the person's personal web pages.
-
Implementation of binarized neural networks immune to device variation and voltage drop employing resistive random access memory bridges and capacitive neuronsIn: Communications Engineering , Vol. 3 2024, 80DOI (Open Access)
-
Powering AI at the edge: A robust, memristor-based binarized neural network with near-memory computing and miniaturized solar cellIn: Nature Communications , Vol. 15 2024, 741DOI (Open Access)
-
Topological magnetic and ferroelectric systems for reservoir computingIn: Nature Reviews Physics , Vol. 6 2024, Nr. 7, pp. 455 – 462
-
Bringing uncertainty quantification to the extreme-edge with memristor-based Bayesian neural networksIn: Nature Communications , Vol. 14 2023, Nr. 1, 7530DOI (Open Access)
-
Model of the Weak Reset Process in HfOₓ Resistive Memory for Deep Learning FrameworksIn: IEEE Transactions on Electron Devices (T-ED) , Vol. 68 2021, Nr. 10, pp. 4925 – 4932DOI, Online Full Text (Open Access)
-
Quantum-Classical Simulation of Molecular Motors Driven only by LightIn: The Journal of Physical Chemistry Letters , Vol. 12 2021, Nr. 23, pp. 5512 – 5518DOI (Open Access)
-
Evolution of the magnetoresistance lineshape with temperature and electric field across Nb-doped SrTiO₃ interfaceIn: Applied Physics Letters (APL) , Vol. 112 2018, Nr. 18, 182405DOI (Open Access)
-
Binary ReRAM-based BNN first-layer implementationIn: 2023 Design, Automation & Test in Europe Conference & Exhibition (DATE): Proceedings / Design, Automation & Test in Europe Conference & Exhibition (DATE), 17-19 April 2023, Antwerp, Belgium / Institute of Electrical and Electronics Engineers (Eds.) 2023DOI, Online Full Text (Open Access)
-
CAPC : A Configurable Analog Pop-Count Circuit for Near-Memory Binary Neural NetworksIn: 2021 IEEE International Midwest Symposium on Circuits and Systems (MWSCAS): on-line proceedings / IEEE International Midwest Symposium on Circuits and Systems (MWSCAS), August 9 – 11, 2021, East Lansing, Michigan, USA, online / Institute of Electrical and Electronics Engineers (Eds.) 2021, pp. 158 – 161DOI, Online Full Text (Open Access)
-
Low-Overhead Implementation of Binarized Neural Networks Employing Robust 2T2R Resistive RAM BridgesIn: ESSCIRC 2021: IEEE 47th European Solid State Circuits Conference (ESSCIRC) ; on-line proceedings / IEEE 47th European Solid State Circuits Conference (ESSCIRC), September 6 – 9, 2021, online / Institute of Electrical and Electronics Engineers (Eds.) 2021, pp. 83 – 86DOI, Online Full Text (Open Access)
-
Embracing Imperfections : Hardware-compatible Neural Networks for Neuromorphic ComputingParis 2023
Journal articles
Book articles / Proceedings papers
Thesis
Advancements in the 21st century are driven by innovations in both materials science and artificial intelligence, fueling progress across technology and industry. My research work lies at the intersection of these fields. I explore in-materio unconventional computing, specifically leveraging the unique physics of nanomagnetic systems (see Figure).
Topological magnetic and ferroelectric systems for reservoir computing
K. Everschor-Sitte, A. Majumdar, K. Wolk, and D. Meier; Nature Review Physics 6, 455 (2024).
Additionally, I apply machine learning to investigate and optimize material properties, aiming to harness these functionalities for practical applications. Ultimately, my goal is to push forward our understanding of materials and contribute to more efficient, sustainable computing technologies.