Zobrazeno 1 - 10
of 139
pro vyhledávání: '"Locatello, Francesco"'
Autor:
Basile, Lorenzo, Maiorca, Valentino, Bortolussi, Luca, Rodolà, Emanuele, Locatello, Francesco
When examined through the lens of their residual streams, a puzzling property emerges in transformer networks: residual contributions (e.g., attention heads) sometimes specialize in specific tasks or input attributes. In this paper, we analyze this p
Externí odkaz:
http://arxiv.org/abs/2411.00246
We consider the linear causal representation learning setting where we observe a linear mixing of $d$ unknown latent factors, which follow a linear structural causal model. Recent work has shown that it is possible to recover the latent factors as we
Externí odkaz:
http://arxiv.org/abs/2410.24059
We propose Scalable Mechanistic Neural Network (S-MNN), an enhanced neural network framework designed for scientific machine learning applications involving long temporal sequences. By reformulating the original Mechanistic Neural Network (MNN) (Perv
Externí odkaz:
http://arxiv.org/abs/2410.06074
Deep learning systems deployed in real-world applications often encounter data that is different from their in-distribution (ID). A reliable system should ideally abstain from making decisions in this out-of-distribution (OOD) setting. Existing state
Externí odkaz:
http://arxiv.org/abs/2410.04525
Autor:
Demirel, Berker, Kong, Lingjing, Zhang, Kun, Karaletsos, Theofanis, Mendler-Dünner, Celestine, Locatello, Francesco
With the widespread deployment of deep learning models, they influence their environment in various ways. The induced distribution shifts can lead to unexpected performance degradation in deployed models. Existing methods to anticipate performativity
Externí odkaz:
http://arxiv.org/abs/2410.04499
Causal representation learning aims at recovering latent causal variables from high-dimensional observations to solve causal downstream tasks, such as predicting the effect of new interventions or more robust classification. A plethora of methods hav
Externí odkaz:
http://arxiv.org/abs/2409.02772
Autor:
Pariza, Valentinos, Salehi, Mohammadreza, Burghouts, Gertjan, Locatello, Francesco, Asano, Yuki M.
We propose sorting patch representations across views as a novel self-supervised learning signal to improve pretrained representations. To this end, we introduce NeCo: Patch Neighbor Consistency, a novel training loss that enforces patch-level neares
Externí odkaz:
http://arxiv.org/abs/2408.11054
Autor:
Montagna, Francesco, Faller, Philipp M., Bloebaum, Patrick, Kirschbaum, Elke, Locatello, Francesco
Causal discovery from observational data holds great promise, but existing methods rely on strong assumptions about the underlying causal structure, often requiring full observability of all relevant variables. We tackle these challenges by leveragin
Externí odkaz:
http://arxiv.org/abs/2407.18755
The emergence of similar representations between independently trained neural models has sparked significant interest in the representation learning community, leading to the development of various methods to obtain communication between latent space
Externí odkaz:
http://arxiv.org/abs/2406.15057
Neural models learn data representations that lie on low-dimensional manifolds, yet modeling the relation between these representational spaces is an ongoing challenge. By integrating spectral geometry principles into neural modeling, we show that th
Externí odkaz:
http://arxiv.org/abs/2406.14183