Prospective students

SOCAL Lab develops statistically principled methods where standard ML pipelines fail: decentralized and privacy-sensitive data, long-tailed regimes, and LLM accountability. We consider motivated Ph.D., M.S., and undergraduate researchers who combine careful theory with clear scientific communication and are prepared for a selective inquiry process.

분산·프라이버시, 불균형·꼬리 구간, LLM 책임성 등에서 통계적으로 엄밀한 방법을 개발합니다. 선별적 문의 절차에 응할 준비가 된 연구자를 검토합니다.

1. Federated learning (data privacy)

Collaborative learning when raw records cannot be centralized, with emphasis on statistical behavior under client heterogeneity and privacy constraints.

• Federated optimization and aggregation under non-IID partitions and distribution shift.
• Privacy-preserving federated protocols (e.g., differential privacy, secure aggregation, and risk-aware design) and their interaction with model utility.
• Communication- and computation-aware federated training under realistic deployment constraints (dropouts, stragglers).

2. Synthetic data

Generation and augmentation of tabular structured data when direct release of sensitive spreadsheets is restricted, including federated and differentially private synthesis.

• High-fidelity tabular synthesis with semantic or structural constraints (including LLM-assisted constraint discovery where appropriate).
• Federated or cross-party synthesis preserving correlations while limiting disclosure of raw private tables.
• Augmentation and resampling to improve coverage in low-density / tail regions, with evaluation beyond headline averages.

3. AI alignment (LLM)

Evaluation- and governance-oriented work on large language models in institutional settings, emphasizing measurable risk and accountability.

• Privacy-aware evaluation and auditing workflows for deployed or procured LLM systems.
• Security, defense, and governance perspectives on heterogeneous LLM deployments (structured surveys and taxonomies).
• Policy–implementation gaps in public-sector technology procurement and post-deployment accountability.

• Mindset and conduct. Members should be self-motivated and determined: research progress depends on sustained initiative through ambiguous phases. Communication skills matter—clear writing and spoken explanation of assumptions, limitations, and findings are part of the research product.
• Technical preparation. Strong probability and statistics and scientific programming (e.g., Python) are required for most projects. Machine learning, numerical linear algebra / optimization, and deep learning experience is beneficial by thread. You should be comfortable reading primary literature and implementing research ideas independently.
• 기술 준비. 확률·통계, Python 등 프로그래밍 필수. 기계학습·최적화·딥러닝 경험 권장. 논문 독해 및 독립적 구현 능력.
• Collaboration. Collegial conduct, and respect for privacy and data-use agreements in applied work.

Support depends on fit and degree level. Eligible members may receive Brain Korea (BK) scholarship support and PI personal research funding (tuition and living on funded lines, per university and grant rules). PI personal research funding backs research assistantships on active projects—not a guarantee for all applicants.

적합성·학위에 따라 지원. BK 장학금 및 교수 개인 연구비(규정 범위). 교수 개인 연구비는 활성 과제 RA 등에 사용되며 전원 보장이 아닙니다.