Contrastive language-image pretraining and latent diffusion models for neural decoding of visua[]

Organisation/Company: Politecnico di Milano Research Field: Engineering Researcher Profile: Recognised Researcher (R2), Leading Researcher (R4), First Stage Researcher (R1), Established Researcher (R3) Country: Italy Application Deadline: 17 Jan 2025 - 23:59 (UTC) Type of Contract: To be defined Job Status: Not Applicable Is the job funded through the EU Research Framework Programme? Not funded by a EU programme Is the Job related to staff position within a Research Infrastructure? No Offer Description One of the potential applications of new paradigms of artificial intelligence based on deep learning is the understanding of what we see. Two main techniques are investigated: Contrastive Language-Image Pre-Training (CLIP) and latent diffusion models. CLIP connects images with their textual descriptions, helping to interpret brain activity. Latent diffusion models, on the other hand, are used to reconstruct images from noise, creating a sort of "reverse translation" of neural activity into images. Integrating the two approaches could be the key to reconstructing what we are looking at by analyzing our brains. Qualifications and Selection Process Relevance of qualifications with the research programme object of selection Consistency of the candidate's overall profile with respect to the content of the research programme for selection Close relevance of publications submitted with the research programme object of selection Interview aimed at ascertaining the candidate's aptitude for the research object of the selection Eligibility of fellows: Country/ies of residence: All Nationality/ies: J-18808-Ljbffr

Location: provincia-di-milano, IT

Posted Date: 1/22/2025