Weight space learning on Implicit Neural Representations

Italian Ministry of Education, University and Research; Italia; In loco; EUR 30.000 - 42.000; Tempo pieno

Descrizione del lavoro

Alma Mater Studiorum - University of Bologna, Dipartimento di Informatica, Scienza e Ingegneria invites applications for a researcher role to study the state of the art on implicit neural representations (INRs) and metanetworks. The project involves developing specifications for INR datasets, shared initializations, and permutation symmetries in neural networks.

You will work on INR alignment algorithms, interpolation optimization, and the training and validation of metanetworks, contributing to

Mansioni

• Study the state of the art on INRs (Implicit Neural Representations) and metanetworks for INR processing.
• Define specifications for building datasets of INRs representing images and 3D data with shared and arbitrary initialization.
• Develop INR alignment algorithms and interpolation optimization procedures.
• Train and validate metanetworks for INR processing.

Descrizione del lavoro

ALMA MATER STUDIORUM - UNIVERSITA' DI BOLOGNA - - DIPARTIMENTO DI INFORMATICA - SCIENZA E INGEGNERIA

Organisation/Company ALMA MATER STUDIORUM - UNIVERSITA' DI BOLOGNA - - DIPARTIMENTO DI INFORMATICA - SCIENZA E INGEGNERIA Research Field Engineering » Computer engineering Researcher Profile Recognised Researcher (R2) Leading Researcher (R4) First Stage Researcher (R1) Established Researcher (R3) Application Deadline 21 Oct 2026 - 23:59 (UTC) Country Italy 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

Study of the state of the art on INRs (Implicit Neural Representations), metanetworks for INR processing, and permutation symmetries of neural networks. Definition of specifications for building datasets of INRs representing images and 3D data, with shared and arbitrary initialization. Development of INR alignment algorithms and of interpolability optimization procedures. Training and validation of metanetworks for INR processing.

• Italy

Eligibility of fellows: country/ies of residence:

• All

Eligibility of fellows: nationality/ies:

Number of offers available 1 Company/Institute ALMA MATER STUDIORUM - UNIVERSITA' DI BOLOGNA - - DIPARTIMENTO DI INFORMATICA - SCIENZA E INGEGNERIA Country Italy