Skip to content

Publications & Research

Thesis

Meta Modeling For AI Solutions Based on Existing AI Configurations M.Sc. Thesis · Yeditepe University · In Progress

Developing a system that extracts structured metadata from AI/ML research papers and recommends optimal model configurations through multi-LLM annotation, QLoRA fine-tuning, and calibrated active learning with vector search.


Papers

Compact Monocular Depth Estimation with Synthetic Stereo Generation

In Preparation

Investigates compact architectures for monocular depth estimation combined with synthetic stereo pair generation for improved stereo matching. Benchmarks multiple architectures and loss functions across datasets.

Project details


Research Interests

  • Meta-Learning & Few-Shot Learning — Learning to learn from limited data
  • Large Language Models & NLP — Fine-tuning, prompt engineering, and evaluation
  • Deep Reinforcement Learning — Policy optimization and model-based methods
  • High-Performance ML Systems — GPU-accelerated training and inference at scale