Postdoc: Multimodal AI for Physical & Natural Sciences

Aretè & Cocchi Technology; Italia; In loco; EUR 35.000 - 45.000; Tempo pieno

Vantaggi offerti da questo lavoro

Private health care

Flexible working time

Staff discounts

Descrizione del lavoro

IIT's Artificial Intelligence for Good (AIGO) research unit invites applications for a postdoctoral position focused on multimodal and generative AI for physical and natural sciences. The postdoc will work in a multicultural team with senior and junior scientists to advance AI methodologies and their applications to diverse domains.

The role emphasizes exploration of hybrid data-driven and physics-informed approaches, with an emphasis on self-supervised and unsupervised learning, and close

Competenze

• PhD in Machine Learning, Computer Vision, Computer Science, Physics, Engineering, Mathematics or related areas.
• Documented expertise in Machine Learning/Deep Learning with multimodal learning.
• Deep generative models (GANs, diffusion, encoder-decoder, Flow Matching).
• ML/CV approaches with emphasis on modern DL models (GNNs, Transformers).
• Strong programming ability (Python) with hands-on experience in PyTorch/TensorFlow.

Mansioni

• Pursue research in interdisciplinary topics within AIGO and IIT flagship activities.
• Supervise PhD students and contribute to mentoring.
• Publish in major conferences/journals.
• Identify and prepare funding opportunities for national and international grants.

Conoscenze

Machine Learning

Computer Vision

Multimodal Learning

Deep Learning

Python

PyTorch

Transformers

Research Publication

Formazione

PhD in Machine Learning/CS/Physics/Engineering

Strumenti

PyTorch

TensorFlow

HPC frameworks

Descrizione del lavoro

IIT's Artificial Intelligence for Good (AIGO) research unit invites applications for a postdoctoral position focused on multimodal and generative AI for physical and natural sciences. The postdoc will work in a multicultural team with senior and junior scientists to advance AI methodologies and their applications to diverse domains.

The role emphasizes exploration of hybrid data-driven and physics-informed approaches, with an emphasis on self-supervised and unsupervised learning, and close