Research Intern on Generative and Protective AI for Content Creation
Sonyjapan
- Tokyo, Japan
- Internship
- Intern/New Grad
- Master's degree
- Posted
Skills
Required
- generative AI
- deep generative modeling
- computer vision
- audio signal processing
- frameworks
- machine learning
- PyTorch
- visualization
- TensorBoard
- CVPR
Preferred
- algorithms
- motion
- data attribution
- coding
Sony AI is seeking research interns who are passionate about ML-based technologies that are useful in movie, game, and music creation, as well as technologies for the ethical operations of generative AIs. Our mission is to research and develop technologies for various Sony Group products and for scientific publication. The technologies developed during the internship will have the potential to be applied in film, game, and music production across entertainment studios worldwide.
Responsibilities
- As a research intern, you will investigate and apply novel algorithms related to: 1) The generation and editing of video, sound, and 3D visual geometry (including 3D human motion).
- The analysis and alleviation of ethical flaws in generative models, including techniques for memorization detection and mitigation, concept erasure, and data attribution.
- Your goal will be to publish your findings in a top-tier conference.
- You are expected to be self-motivated and to implement innovative ideas using your research, coding, and problem-solving skills.
- You will receive support from internal scientists and engineers in your efforts.
Requirements
- Qualifications Master’s degree in Computer Science, Electrical Engineering, Applied Mathematics, or related fields.
- Proven knowledge and expertise in generative AI applications, including deep generative modeling, computer vision, and audio signal processing.
- Strong analytical and programming skills in deep learning using frameworks and tools for machine learning (e.g., PyTorch) and visualization (e.g., TensorBoard).
- Experience in research communities, including published papers at top-tier conferences such as CVPR, ICCV, ECCV, NeurIPS, ICLR, and ICML.
- Excellent communication and presentation skills.
Preferred
- Qualifications Currently enrolled in a relevant Ph.D. program in fields such as Computer Science, Electrical Engineering, Applied Mathematics, or related disciplines.
Working
Location Location flexible (Tokyo, NYC, remote) Required Skills: Optional Skills:
