Autoregressive models have achieved impressive results over a wide range of domains in terms of generation quality and downstream task performance. In the continuous domain, a key factor behind this success is the usage of quantized latent spaces (e.g., obtained via VQ-VAE autoencoders), which al low for dimensionality reduction and faster inference times. However, using existing pre-trained models to perform new non-trivial tasks is difficult since it requires additional fine tuning or extensive training to elicit prompting. This paper introduces LASS as a way to perform vector-quantized La tent Autoregressive Source Separation (i.e., de-mixing an in put signal into its constituent sources) without requiring ad ditional gradient-based optimization or modifications of ex isting models. Our separation method relies on the Bayesian formulation in which the autoregressive models are the pri ors, and a discrete (non-parametric) likelihood function is constructed by performing frequency counts over latent sums of addend tokens. We test our method on images and au dio with several sampling strategies (e.g., ancestral, beam search) showing competitive results with existing approaches in terms of separation quality while offering at the same time significant speedups in terms of inference time and scalability to higher dimensional data.

Latent Autoregressive Source Separation / Postolache, Emilian; Mariani, Giorgio; Mancusi, Michele; Santilli, Andrea; Cosmo, Luca; Rodola', Emanuele. - (2023). (Intervento presentato al convegno The Thirty-Seventh AAAI Conference on Artificial Intelligence tenutosi a Washington DC, USA).

Latent Autoregressive Source Separation

Emilian Postolache;Giorgio Mariani;Michele Mancusi;Andrea Santilli;Emanuele Rodola`
2023

Abstract

Autoregressive models have achieved impressive results over a wide range of domains in terms of generation quality and downstream task performance. In the continuous domain, a key factor behind this success is the usage of quantized latent spaces (e.g., obtained via VQ-VAE autoencoders), which al low for dimensionality reduction and faster inference times. However, using existing pre-trained models to perform new non-trivial tasks is difficult since it requires additional fine tuning or extensive training to elicit prompting. This paper introduces LASS as a way to perform vector-quantized La tent Autoregressive Source Separation (i.e., de-mixing an in put signal into its constituent sources) without requiring ad ditional gradient-based optimization or modifications of ex isting models. Our separation method relies on the Bayesian formulation in which the autoregressive models are the pri ors, and a discrete (non-parametric) likelihood function is constructed by performing frequency counts over latent sums of addend tokens. We test our method on images and au dio with several sampling strategies (e.g., ancestral, beam search) showing competitive results with existing approaches in terms of separation quality while offering at the same time significant speedups in terms of inference time and scalability to higher dimensional data.
2023
The Thirty-Seventh AAAI Conference on Artificial Intelligence
Autoregressive models; quantized latent spaces; non-trivial tasks; vector-quantized Latent Autoregressive Source Separation
04 Pubblicazione in atti di convegno::04b Atto di convegno in volume
Latent Autoregressive Source Separation / Postolache, Emilian; Mariani, Giorgio; Mancusi, Michele; Santilli, Andrea; Cosmo, Luca; Rodola', Emanuele. - (2023). (Intervento presentato al convegno The Thirty-Seventh AAAI Conference on Artificial Intelligence tenutosi a Washington DC, USA).
File allegati a questo prodotto
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11573/1672044
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact