DPVI: A Dynamic-Weight Particle-Based Variational Inference Framework

Abstract

Deep generative models, such as Variational Autoencoders (VAEs) and Generative Adversarial Networks (GANs), have been employed widely in computational creativity research. However, such models discourage out-of-distribution generation to avoid spurious sample generation, thereby limiting their creativity. Thus, incorporating research on human creativity into generative deep learning techniques presents an opportunity to make their outputs more compelling and human-like. As we see the emergence of generative models directed toward creativity research, a need for machine learning-based surrogate metrics to characterize creative output from these models is imperative. We propose group-based subset scanning to identify, quantify, and characterize creative processes by detecting a subset of anomalous node-activations in the hidden layers of the generative models. Our experiments on the standard image benchmarks and their ``creatively generated'' variants reveal that the proposed subset scores distribution is more useful for detecting novelty in creative processes in the activation space rather than the pixel space. Further, we found that creative samples generate larger subsets of anomalies than normal or non-creative samples across datasets. The node activations highlighted during the creative decoding process are different from those responsible for the normal sample generation. Lastly, we assess if the images from the subsets selected by our method were also found creative by human evaluators, presenting a link between creativity perception in humans and node activations within deep neural nets.

GAD-PVI: A General Accelerated Dynamic-Weight Particle-Based Variational Inference Framework

“…Recently, the Wasserstein-Fisher-Rao (WFR) Flow has been used to derive effective dynamic weight adjustment approaches to mitigate the fixed-weight restriction of ParVIs (Zhang et al 2022). The inverse of WFR metric tensor is…”

“…(9) According to the ode (9), Zhang et al (2022) derive two dynamical weight-adjustment scheme and propose the Dynamic-Weight Particle-Based Variational Inference (DPVI) framework, which is compatible with several dissimilarity functionals and associated smoothing approaches.…”

“…Particle-based Variational Inference (ParVI) methods have gained significant attention in the Bayesian inference literature owing to their effectiveness in providing approximations of the target posterior distribution π (Liu and Wang 2016;Zhu, Liu, and Zhu 2020;Dong et al 2020;Shen, Heinonen, and Kaski 2021;Dong et al 2021;Zhang et al 2022;Li et al 2023). The essence of ParVI lies in deterministically evolving a system of finite weighted particles by simulating the probability space gradient flow of certain dissimilarity functional F (µ) := D(µ|π) vanishing at µ = π (Liu et al 2019).…”

“…• Dynamic weight adjustment. Delving into the orthogonality structure of the Wasserstein-Fisher-Rao(WFR) space, Zhang et al (2022) developed the first dynamicweight ParVI (DPVI) methods. Specifically, they derive effective dynamical weight adjustment approaches by mimicing the reaction variational step in a JKO splitting scheme of first-order WFR gradient flow (Gallouët and Monsaingeon 2017;Rotskoff et al 2019).…”

“…Specifically, they derive effective dynamical weight adjustment approaches by mimicing the reaction variational step in a JKO splitting scheme of first-order WFR gradient flow (Gallouët and Monsaingeon 2017;Rotskoff et al 2019). Compared with the commonly used fixed weight strategy, these dynamical weight adjustment schemes usually lead to less approximation error, especially when the number of particles is limited (Zhang et al 2022).…”