Projects:

Visualising Quantum Time Evolution using Qiskit 
GitHub Repo
Project Page
The Split operator Fourier transform algorithm is used to simulate the phenomenon of Quantum tunnelling using Quantum computers, which inherently use quantum physics. Hamiltonian of the system is broken down and split into an infinite amount of smaller steps alternating between kinetic and potential propagators using the Lie‑Trotter product formula. The wave function initialized is discretized into 2𝑁 grid points where N is the number of qubits (excluding ancillary) used in the circuit. Following this, the potential operator, the Quantum Fourier Transform (QFT), the kinetic operator, and the inverse QFT, is applied for the desired number of iterations. The circuit is then measured and represented as a histogram, which gives us a clear visualization of the Quantum‑tunneling phenomenon. After understanding the fundamentals of using the split operator approach to simulate one particle moving, my motivation is to investigate particle interactions and chemical dynamics, which has tremendous research applications in domains such as drug development, condensed matter physics, and material science.

Computation of periodic materials using VQE 
GitHub Repo
The ground state energy for covalent organic framework (sp2C‑COF and sp2N‑COF) molecules with the Lieb structure is calculated. This is done by using the VQE algorithm on the Qamuy quantum computing software. The idea was to perform the VQE calculations on the monomers of one of the Lieb lattices. Aromatic rings make up for most of the lattice (as substituted pyrene) and along with the ‘ethyl cyanide’ moiety. Since the monomers were too large, the units of the monomers themselves (benzene and isopropyl‑cyanide) were taken up for simulation. The reported calculations were done at the DFT level which is always accompanied by errors arising from the functional approximations. In this project, the calculations using the VQE algorithm (optimization and then single point) for the benzene and isopropyl‑cyanide units is compared with the results to MP2 level observations using the 6‑311G* basis sets. Building on this project, it is possible to calculate the binding energy of the sp2C‑COF Lieb lattice within a certain degree of approximation and this could potentially provide promising results aiding in the understanding of the real Lieb lattices.

Entangled Souls
GitHub Repo
Game Page
Entangled Souls is a story‑driven 3D puzzle room video game which utilizes horror and quantum elements in the design. Due to the inter‑disciplinary nature of quantum‑game development, each of our team member had a unique role and contribution responsibility. I was involved in the project as the ’Quantum Expert’ and contributed as the lead puzzle designer. ’Entangled Souls’ game takes inspiration from quantum and uses the theme in the narrative and gameplay. Quantum physics is utilized in puzzles and game mechanics.

Quantum Image Processing using FRQI 
GitHub Repo
Quantum Image Processing is a fast growing field within the quantum computing space. A simple 4x4 greyscale image is encoded, rotated and translated inside a 8x8 grid. The program utilizes the Flexible Representation of Quantum Images (FRQI). The rotations and translations of the grayscale checkerboard image matrix are processed on quantum computers. The results from different quantum hardware and quantum simulators are compared. Contrary to quantum simulators, the effect of noise and errors is clearly observed in results from quantum hardware.

Solving binding energies for the ground state protease in SARS‑Cov‑2
GitHub Repo
A protease is an enzyme that works as a ’hammer’, breaking down proteins into smaller amino acids and polypeptides. Covid‑19 contains a protease called glutamine. By blocking the protease’s receptors, we effectively halts the ability of the virus to spread within its host. This can be done by reacting a protease inhibitor with the protease to stop the chemical process from occurring. Current quantum hardware is not powerful enough to simulate complete macromolecules, and hence, the glutamine molecule is simplified to create a a 'toy protease', to make the problem simpler and prove the effectiveness of quantum computers. While only a single carbon atom bounded to the glutamine molecule is simulated here, it demonstrates the ability of quantum computational chemistry in current chemical problems. By demonstrating these results using VQE, we could potentially achieve speedups in the development of novel vaccines to combat the different virus mutations.

Introduction to Quantum Chemistry
GitHub Repo
As there are not many introductory resources available on the application of Quantum Chemistry in Quantum Computation, this is a set of introductory Jupyter Notebooks based on a pedagogical approach to Quantum Chemistry, which can be used for teaching QChemistry to a wide set of audiences with minimal prerequisites. The complete resource has been developed and remains OpenSource with the aim of helping anyone interested in learning the subject.

Deep learning for compression of classical data in quantum computing
GitHub Repo
For any project of quantum computing or quantum machine learning, data compression of classical data is a must in a NISQ era due to the limitation on the number of stable qubits available to us. Based on universal approximation theorems, deep learning (DL) could approximate any function. It is hypothesized that DL would learn the optimized parameters to compress the classical data for QC/QML. Information loss would be minimized during the data compression with deep learning. Each quantum circuit would be suitable for different data compression models (both hyperparameters and parameters) and one could train and design different DL data compression model structures for several quantum circuits. Multitask Learning is used on QC/QML to simultaneously minimize the loss of autoencoder and loss of performance of QC/QML. Finally, the complete tool has been made opensource so that one could perform end-to-end training with inputting classical data without the need of complicated coding.

Quantum Defect Analyser
GitHub Repo
Quantum Neural Networks (QNN) are parameterized quantum circuits which act like linear methods in quantum feature space and suitable case for using quantum kernels and in turn can be optimized using Quantum Kernel Alignment (QKA). QKA is an iterative quantum-classical algorithm, in which quantum hardware is used to execute parametrized quantum circuits for evaluating the quantum kernel matrices with QKE, while a classical optimizer tunes the parameters of those circuits to maximize the alignment. Iterative algorithms of this type can be slow due to latency between the quantum and classical calculations. In this project, QKA is used for optimizing Quantum Neural Network, which processes images from Surface Crack Detection Kaggle dataset. Potentially, it can be scaled up & used as a module in Automated Quality Management in assembly line of various Manufacturing Factory units.

The Quantum Story
Project Page
A short and fun to read educational resource for teaching the fundamentals of Quantum Computing. Keeping in mind our primary target audience as the student body of Higher Secondary level, a conversational style of teaching is implemented to make the study material as easy to understand as possible. Intelligent use of real life examples help relate the different disciplines like mathematics and science to formulate and understand the fundamentals of QC in this quasi-storyboard format of education.

Quaze
GitHub Repo
Game Page
Quaze is a simple puzzle game developed using JS, HTML and CSS. The objective of the game is to reach the exit of the maze and have the qubit in the desired target state, as indicated at the end of the maze. By using the fundamental concepts of QC like transformation of qubits, a challenging twist is given to the traditional game of ’Maze’, thus giving rise to 'Quaze'.

Modelling of Positron‑Impact Direct Ionisation of Molecules
Project Report
The study of positron-impact ionisation is not very well established as that of electron‑impact ionisation. The problem of calculating the cross sections of positron‑impact ionisation is still prominent. In this project, the BEB method for electrons is extended to include positron‑impact ionisation cross sections after some mathematical modifications to the model. The new model developed gives very good results for the few molecules examined here. It is encouraged that the model be verified for various other molecules as well in order to confirm its validity.

Computational Approach to Quantum Mechanical Systems
Project Report
Practical project of realising and simulating various simple quantum systems like Linear Harmonic Oscillator, Infinite Square Well potential, and 1D Hydrogen atom using computational numerical methods on SciLab.

Gravitational Wave Data Analysis
Project Report
Three different gravitational wave events are analyzed with a simple algorithm that can be implemented to any of the LIGO datasets to obtain the chirp (with minor modifications). The detection of the chirp is followed by parameter estimation. Open-source code of the bilby module has been used to obtain the important parameters of the three events. In the first case, the pre-published result of the first event, GW150914, is reproduced, the event’s waveform described, a matched filtering and parameter estimation performed, finally plotting a corner plot of the result. In the second case, the parameter estimation of one event each from O1, O2 and O3a dataset is done by reducing the dimensionality of the problem, thus, limiting the number of priors to be calculated. Finally, the results of the published data and the data obtained from our parameter estimation is compared and the inferences recorded.

- - Vardaan Sahgal - -

‌