Speaker: Dr. Himangshu Prabal Goswami
Assistant Professor, Department of Chemistry,
University of Gauhati, Guwahati 781014,
Assam India
Title: “Learning coherences from nonequilibrium fluctuations
in quantum heat engines.”
Day and Date: Monday, November 28, 2022
Time: 4.00 pm.
Venue: Room no. 350, Chemistry Department
Second floor, Annex
------------------------------------------------------------------------
Hosted by Prof. Rajarshi Chakrabarti & Prof. Srinivasan Ramakrishnan
Abstract Traditional or classical thermal engines absorb heat to produce useful
mechanical work. Quantum heat engines in addition to having traditional
properties are also affected by coherences and nonequilibrium fluctuations
that arise due to the interactions between a quantized system and the
quantum properties of the thermal baths [1]. The dependence of higher order
power fluctuations on the quantum coherences however is nontrivial owing to
the lack of knowledge of the relationship between these quantities despite
both being experimentally measurable [2]. Usually, for a designed quantum
engine, the coherences are rendered fixed while measuring the fluctuations.
The other system parameters like coupling and temperatures act as control
knobs to obtain desired values of fluctuations [2]. But these change the
coherences creating an unknown mapping between observed and desired values.
In this talk, I shall introduce a popular quantum heat engine where
coherences are known to optimize the power beyond classical values [1,2]. I
shall also show how one theoretically evaluates the experimentally observed
fluctuations using a full counting statistical approach. Further, I shall
show how machine learning algorithms based on gradient boosting and
artificial neural networks allow us to use the information obtained from
the fluctuations to predict the coherence values as well as other quantum
thermodynamic observables and thermodynamic uncertainty relations.