PhD Candidate in Explainable AI for Synthetic Media

NTNU; Norway; On-site; NOK 522,000 - 638,000; Full time

Benefits offered by this job

Free Norwegian language training at a.

Job summary

NTNU seeks a PhD Candidate in Explainable AI for Synthetic Media Analysis at the Gjøvik campus. The PhD candidate will develop explainable ML methods to analyze authentic, synthetic, and manipulated images or videos, combining performance with interpretability and uncertainty estimation.

The role offers doctoral training, collaboration, and access to leading research environments. The position requires a Master’s degree with a strong research track record, proficiency in Python and PyTorch, and

Qualifications

• Master's degree or equivalent in computer science, artificial intelligence, machine learning, computer vision, or signal processing with a substantial independent project equivalent to a master's thesis.
• Strong academic background with average grade of B or better on NTNU scale; exceptional candidates may be considered with relevant work/publications.
• Good knowledge of machine learning and deep learning, with evidence of experimental methodology.
• Strong programming skills in Python and experience with PyTorch or TensorFlow.
• Good written and oral English communication.
• Experience in computer vision, video analysis, self-supervised learning, vision transformers, or multimodal learning.
• Experience with pretrained representation models, transfer learning, parameter-efficient adaptation, or representation probing.
• Knowledge of media forensics, synthetic-content analysis, manipulation localization, or related authenticity problems.
• Knowledge of explainable AI, uncertainty estimation, calibration, OOD, adversarial robustness, or causal analysis.
• Experience with causal inference, counterfactual explanations, or uncertainty quantification in deep learning.
• Evidence of high-quality scientific writing, publications, a strong master's thesis, research software, or relevant open-source contributions.
• Prior experience working across geographically distributed teams.

Responsibilities

• Develop machine-learning methods for detecting, localizing, or characterizing synthetic and manipulated visual content.
• Study representations learned by large pretrained visual or multimodal encoders, including probing, adaptation, fusion, and efficient fine-tuning strategies.
• Design explanation methods that connect predictions to meaningful spatial, temporal, frequency-domain, semantic, or example-based evidence.
• Evaluate robustness and transfer to unseen datasets, content sources, and manipulation processes.
• Investigate uncertainty, calibration, and reliability measures that support responsible human decision-making.
• Develop reproducible benchmarks and evaluation protocols for predictive quality, explanation fidelity and stability, and computational efficiency.
• Publish results in peer-reviewed venues and present at conferences and with interdisciplinary audiences.
• Participate in doctoral training and collaboration with relevant academic and external research environments.

Skills

Python programming

PyTorch

Deep learning

English proficiency

Academic writing

Education

Master's degree in CS/AI/ML/CV/Signal Processing

Tools

Git

TensorFlow

Docker

Job description

NTNU seeks a PhD Candidate in Explainable AI for Synthetic Media Analysis at the Gjøvik campus. The PhD candidate will develop explainable ML methods to analyze authentic, synthetic, and manipulated images or videos, combining performance with interpretability and uncertainty estimation.

The role offers doctoral training, collaboration, and access to leading research environments. The position requires a Master’s degree with a strong research track record, proficiency in Python and PyTorch, and