Publication

Abstract : A deep neural network is a branch of machine learning that is capable of learning and representing complex patterns from a dataset through interconnected multiple layers of neurons. This capability makes it applicable in various fields, such as natural language processing, image processing, and computer vision. Deep learning models show effective performance but face challenges such as complexity and resource demands. On the other hand, quantum machine learning algorithms offer an alternative with potential efficiency compared to their classical counterparts. This paper proposes a Quantum Recurrent Neural Network (QRNN) for natural language processing tasks, which classify text data such as parts of speech, named entity recognition, and sentiment analysis. The proposed method utilizes parameterized quantum circuits that contain the tunable parameters. Our approach uses amplitude encoding to represent classical data into quantum states, partial measurement for label determination, and ancilla qubits to transfer the information from the current state to the next.