Abstract:We consider fully connected and feedforward deep neural networks with dependent and possibly heavy-tailed weights, as introduced in [26], to address limitations of the standard Gaussian prior. It has been proved in [26] that, as the number of nodes in the hidden layers grows large, according to a sequential and ordered limit, the law of the output converges weakly to a Gaussian mixture. In this paper, we study the neural network through the lens of the posterior distribution with a Gaussian likelihood. If the random covariance matrix of the infinite-width limit is positive definite under the prior, we identify the posterior distribution of the output in the wide-width limit according to a sequential regime. Remarkably, we provide mild sufficient conditions to ensure the aforementioned invertibility of the random covariance matrix under the prior, thereby extending the results in [8]. Among our results, we present sufficient conditions on some model parameters (the activation function and the associated Lévy measures) which ensure that the sequential limits are independent of the order. We illustrate our findings with examples and numerical simulations.
| Comments: | 2 figures |
| Subjects: | Machine Learning (stat.ML); Machine Learning (cs.LG); Probability (math.PR) |
| MSC classes: | 60F05, 62F15, 68T07 |
| Cite as: | arXiv:2507.22095 [stat.ML] |
| (or arXiv:2507.22095v5 [stat.ML] for this version) | |
| https://doi.org/10.48550/arXiv.2507.22095 arXiv-issued DOI via DataCite |
Submission history
From: Giovanni Franzina [view email]
[v1]
Tue, 29 Jul 2025 15:54:34 UTC (414 KB)
[v2]
Sat, 20 Sep 2025 14:19:36 UTC (495 KB)
[v3]
Sat, 10 Jan 2026 17:29:31 UTC (426 KB)
[v4]
Wed, 14 Jan 2026 11:50:48 UTC (426 KB)
[v5]
Wed, 13 May 2026 08:41:44 UTC (417 KB)