# Negative Binomial Variational Autoencoders for Overdispersed Latent Modeling

arXiv: https://arxiv.org/abs/2508.05423 · HTML: https://mlsanity.com/paper/2508.05423/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 58 references that are in the map.
- First posted on arXiv 7 Aug 2025.

## Builds on (most cited first, 24 of 24)

- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms](https://mlsanity.com/paper/1708.07747/) (2017)
- [Deep Learning Face Attributes in the Wild](https://mlsanity.com/paper/1411.7766/) (2014)
- [Categorical Reparameterization with Gumbel-Softmax](https://mlsanity.com/paper/1611.01144/) (2016)
- [NVAE: A Deep Hierarchical Variational Autoencoder](https://mlsanity.com/paper/2007.03898/) (2020)
- [Incorporating Learnable Membrane Time Constant to Enhance Learning of Spiking Neural Networks](https://mlsanity.com/paper/2007.05785/) (2020)
- [Going Deeper With Directly-Trained Larger Spiking Neural Networks](https://mlsanity.com/paper/2011.05280/) (2020)
- [Very Deep VAEs Generalize Autoregressive Models and Can Outperform Them on Images](https://mlsanity.com/paper/2011.10650/) (2020)
- [Deep Learning in Spiking Neural Networks](https://mlsanity.com/paper/1804.08150/) (2018)
- [Discrete Variational Autoencoders](https://mlsanity.com/paper/1609.02200/) (2016)
- [Implicit Reparameterization Gradients](https://mlsanity.com/paper/1805.08498/) (2018)
- [Learning Disentangled Joint Continuous and Discrete Representations](https://mlsanity.com/paper/1804.00104/) (2018)
- [SOM-VAE: Interpretable Discrete Representation Learning on Time Series](https://mlsanity.com/paper/1806.02199/) (2018)
- [Fully Spiking Variational Autoencoder](https://mlsanity.com/paper/2110.00375/) (2021)
- [Predictive Coding, Variational Autoencoders, and Biological Connections](https://mlsanity.com/paper/2011.07464/) (2020)
- [Spiking Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2306.17046/) (2023)
- [Latent Diffusion for Neural Spiking Data](https://mlsanity.com/paper/2407.08751/) (2024)
- [Poisson Variational Autoencoder](https://mlsanity.com/paper/2405.14473/) (2024)
- [Spiking-GAN: A Spiking Generative Adversarial Network Using Time-To-First-Spike Coding](https://mlsanity.com/paper/2106.15420/) (2021)
- [Spiking Generative Adversarial Networks With a Neural Network Discriminator: Local Training, Bayesian Models, and Continual Meta-Learning](https://mlsanity.com/paper/2111.01750/) (2021)
- [Spiking Generative Adversarial Network with Attention Scoring Decoding](https://mlsanity.com/paper/2305.10246/) (2023)
- [Spiking-Diffusion: Vector Quantized Discrete Diffusion Model with Spiking Neural Networks](https://mlsanity.com/paper/2308.10187/) (2023)
- [Deterministic Decoding for Discrete Data in Variational Autoencoders](https://mlsanity.com/paper/2003.02174/) (2020)
- [ESVAE: An Efficient Spiking Variational Autoencoder with Reparameterizable Poisson Spiking Sampling](https://mlsanity.com/paper/2310.14839/) (2023)

## Built on by (most cited first, 2 of 2)

- [A hitchhiker's guide to Poisson gradient estimation](https://mlsanity.com/paper/2602.03896/) (2026)
- [Metabolic cost of information processing in Poisson variational autoencoders](https://mlsanity.com/paper/2602.13421/) (2026)

## Abstract (arXiv)

Although artificial neural networks are often described as brain-inspired, their representations typically rely on continuous activations, such as the continuous latent variables in variational autoencoders (VAEs), which limits their biological plausibility compared to the discrete spike-based signaling in real neurons. Extensions like the Poisson VAE introduce discrete count-based latents, but their equal mean-variance assumption fails to capture overdispersion in neural spikes, leading to less expressive and informative representations. To address this, we propose NegBio-VAE, a negative-binomial latent-variable model with a dispersion parameter for flexible spike count modeling. NegBio-VAE preserves interpretability while improving representation quality and training feasibility via novel KL estimation and reparameterization. Experiments on four datasets demonstrate that NegBio-VAE consistently achieves superior reconstruction and generation performance compared to competing single-layer VAE baselines, and yields robust, informative latent representations for downstream tasks. Extensive ablation studies are performed to verify the model's robustness w.r.t. various components. Our code is available at https://github.com/co234/NegBio-VAE.