# On the Joint Minimization of Regularization Loss Functions in Deep Variational Bayesian Methods for Attribute-Controlled Symbolic Music Generation

arXiv: https://arxiv.org/abs/2511.07118 · HTML: https://mlsanity.com/paper/2511.07118/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 25 references that are in the map.
- First posted on arXiv 10 Nov 2025.

## Builds on (most cited first, 9 of 9)

- [Deep Variational Information Bottleneck](https://mlsanity.com/paper/1612.00410/) (2016)
- [Symbolic Music Generation with Diffusion Models](https://mlsanity.com/paper/2103.16091/) (2021)
- [A Hierarchical Latent Vector Model for Learning Long-Term Structure in Music](https://mlsanity.com/paper/1803.05428/) (2018)
- [Latent Constraints: Learning to Generate Conditionally from Unconditional Generative Models](https://mlsanity.com/paper/1711.05772/) (2017)
- [Encoding Musical Style with Transformer Autoencoders](https://mlsanity.com/paper/1912.05537/) (2019)
- [Learning to Groove with Inverse Sequence Transformations](https://mlsanity.com/paper/1905.06118/) (2019)
- [GLSR-VAE: Geodesic Latent Space Regularization for Variational AutoEncoder Architectures](https://mlsanity.com/paper/1707.04588/) (2017)
- [Attri-VAE: attribute-based interpretable representations of medical images with variational autoencoders](https://mlsanity.com/paper/2203.10417/) (2022)
- [Bridging Explicit and Implicit Deep Generative Models via Neural Stein Estimators](https://mlsanity.com/paper/1909.13035/) (2019)

## Abstract (arXiv)

Explicit latent variable models provide a flexible yet powerful framework for data synthesis, enabling controlled manipulation of generative factors. With latent variables drawn from a tractable probability density function that can be further constrained, these models enable continuous and semantically rich exploration of the output space by navigating their latent spaces. Structured latent representations are typically obtained through the joint minimization of regularization loss functions. In variational information bottleneck models, reconstruction loss and Kullback-Leibler Divergence (KLD) are often linearly combined with an auxiliary Attribute-Regularization (AR) loss. However, balancing KLD and AR turns out to be a very delicate matter. When KLD dominates over AR, generative models tend to lack controllability; when AR dominates over KLD, the stochastic encoder is encouraged to violate the standard normal prior. We explore this trade-off in the context of symbolic music generation with explicit control over continuous musical attributes. We show that existing approaches struggle to jointly minimize both regularization objectives, whereas suitable attribute transformations can help achieve both controllability and regularization of the target latent dimensions.