| Title |
Investigation of classification algorithms in quantum computing / |
| Translation of Title |
Klasifikavimo užduotims skirtų kvantinių algoritmų nagrinėjimas. |
| Authors |
Paliulis, Skalvis |
| Full Text |
|
| Pages |
57 |
| Keywords [eng] |
Klasifikavimas, Kvantiniai skaičiavimai, PennyLane, Lyginamoji analizė, Classification, Quantum Computing, PennyLane, Benchmark |
| Abstract [eng] |
This master‘s thesis aims to identify and empirically investigate classification algorithms implementation in quantum computing, focusing on PennyLane framework utility. Due to substantial increase in computing power via quantum computers in recent years, actual application of quantum algorithms and their performance against classical algorithms are relevant. Moreover, given that classical algorithms optimization functions rely on non-convex objective functions, they tend to be stuck within local maxima/minima. With quantum algorithms, we can either utilize quantum bits/superposition/entanglement to represent multiple states as their respective possibilities to escape local maxima or finding global minima. The benchmark results indicate that quantum classification algorithms perform similarly to their classical algorithms counterparts, however are time inefficient and not well-suited for large data sets. There are many similar frameworks for quantum computing. PennyLane is compatible with any gate-based quantum simulator or hardware, whereas others are not (i.e., Qiskit). Compatibility importance cannot be understated. |
| Dissertation Institution |
Vilniaus universitetas. |
| Type |
Master thesis |
| Language |
English |
| Publication date |
2024 |
