Approximate Bayesian Inference via Bitstring Representations

Aleksanteri Sladek, Martin Trapp, Arno Solin
Proceedings of the Forty-first Conference on Uncertainty in Artificial Intelligence, PMLR 286:3939-3957, 2025.

Abstract

The machine learning community has recently put effort into quantized or low-precision arithmetics to scale large models. This paper proposes performing probabilistic inference in the quantized, discrete parameter space created by these representations, effectively enabling us to learn a continuous distribution using discrete parameters. We consider both 2D densities and quantized neural networks, where we introduce a tractable learning approach using probabilistic circuits. This method offers a scalable solution to manage complex distributions and provides clear insights into model behavior. We validate our approach with various models, demonstrating inference efficiency without sacrificing accuracy. This work advances scalable, interpretable machine learning by utilizing discrete approximations for probabilistic computations.

Cite this Paper

BibTeX
@InProceedings{pmlr-v286-sladek25a, title = {Approximate Bayesian Inference via Bitstring Representations}, author = {Sladek, Aleksanteri and Trapp, Martin and Solin, Arno}, booktitle = {Proceedings of the Forty-first Conference on Uncertainty in Artificial Intelligence}, pages = {3939--3957}, year = {2025}, editor = {Chiappa, Silvia and Magliacane, Sara}, volume = {286}, series = {Proceedings of Machine Learning Research}, month = {21--25 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v286/main/assets/sladek25a/sladek25a.pdf}, url = {https://proceedings.mlr.press/v286/sladek25a.html}, abstract = {The machine learning community has recently put effort into quantized or low-precision arithmetics to scale large models. This paper proposes performing probabilistic inference in the quantized, discrete parameter space created by these representations, effectively enabling us to learn a continuous distribution using discrete parameters. We consider both 2D densities and quantized neural networks, where we introduce a tractable learning approach using probabilistic circuits. This method offers a scalable solution to manage complex distributions and provides clear insights into model behavior. We validate our approach with various models, demonstrating inference efficiency without sacrificing accuracy. This work advances scalable, interpretable machine learning by utilizing discrete approximations for probabilistic computations.} }
Endnote
%0 Conference Paper %T Approximate Bayesian Inference via Bitstring Representations %A Aleksanteri Sladek %A Martin Trapp %A Arno Solin %B Proceedings of the Forty-first Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2025 %E Silvia Chiappa %E Sara Magliacane %F pmlr-v286-sladek25a %I PMLR %P 3939--3957 %U https://proceedings.mlr.press/v286/sladek25a.html %V 286 %X The machine learning community has recently put effort into quantized or low-precision arithmetics to scale large models. This paper proposes performing probabilistic inference in the quantized, discrete parameter space created by these representations, effectively enabling us to learn a continuous distribution using discrete parameters. We consider both 2D densities and quantized neural networks, where we introduce a tractable learning approach using probabilistic circuits. This method offers a scalable solution to manage complex distributions and provides clear insights into model behavior. We validate our approach with various models, demonstrating inference efficiency without sacrificing accuracy. This work advances scalable, interpretable machine learning by utilizing discrete approximations for probabilistic computations.
APA
Sladek, A., Trapp, M. & Solin, A.. (2025). Approximate Bayesian Inference via Bitstring Representations. Proceedings of the Forty-first Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 286:3939-3957 Available from https://proceedings.mlr.press/v286/sladek25a.html.

Related Material