Introduces the Quantum Recurrent Unit (QRU), a NISQ-compatible quantum neural network inspired by classical Gated Recurrent Units, achieving constant circuit depth regardless of input sequence length. Validated across multiple benchmarks, QRU matches or outperforms classical networks with a fraction of the parameters — reaching 98.05% MNIST accuracy with just 132 parameters vs. a 27,265-parameter CNN.

Explores how quantum computing is transforming drug discovery and biomedicine, with applications extending across other industries. Introduces new hybrid quantum-classical algorithms that bridge near-term quantum hardware with classical computation, and discusses Korea's national R&D initiatives driving these advances forward.

An exploration of how quantum neural networks can be leveraged to make reinforcement learning more efficient.

Quantum-inspired algorithms, implemented through Simulated Annealing and Quantum Annealing with QUBO-like formulations, effectively tackle the challenge of local minima in high-dimensional solution spaces. This approach enables efficient identification of key transaction features linked to high-risk Bitcoin mixer addresses, significantly reducing machine learning model training time by over 30% while maintaining a strong 91% F1-score. The talk will discuss the methodology, experimental results on real Bitcoin blockchain data, and broader implications for QAI-driven applications in blockchain security, AML/RegTech, and privacy-preserving finance.

Co-presenting with Prof. Ming-Fong Sie in a dual panel on QAI applications in finance and feature selection.

A mathematical framework for constructing basis vectors that achieve maximum entanglement across N-qubit quantum systems.

For the upcoming space exploration era, advanced AI feature extraction/learning becomes increasingly critical for satellite data understanding. Quantum deep network (QUEEN) offers counterpart feature information to be fused with classical ones for better decision-making. Module design philosophy and applications to remote sensing and bioinformatics will be presented.

Covers the computation of analytical gradients in Variational Quantum Algorithms (VQA) — a key technique for efficiently training quantum circuits on near-term NISQ hardware. Also introduces QISCA (Quantum Information Science Club Association), Korea's largest student-led quantum organization: a nationwide coalition of clubs from eight top universities including KAIST, Seoul National University, and Yonsei University. Wonjun Baek, QISCA President, will share the organization's mission to connect students across disciplines through seminars, seasonal schools, lab visits, and annual conferences.

Welcome to SQAI 2026 and opening remarks, followed by a presentation of Austin's vision paper on quantum ultracomputing and the path to Artificial Ultraintelligence. Also participates in the NTUQC-SQAI Student Panel alongside Chao Hsien and Wonjun Baek.

A 50-minute student panel featuring Chao Hsien, Wonjun Baek, and Austin Hua — covering NTUQC's mission and community, QISCA and the broader student quantum ecosystem, and a discussion on quantum AI from the next generation's perspective.