8 citations · 16 across the 19 of their papers we have counts for
Showing 2023 · stat.MLShow all
3 papers · 2 filters
stat.ML2023★ 3 cited
Learning Dynamics in Linear VAE: Posterior Collapse Threshold, Superfluous Latent Space Pitfalls, and Speedup with KL Annealing
Yuma Ichikawa, Koji Hukushima
Variational autoencoders (VAEs) face a notorious problem wherein the variational posterior often aligns closely with the prior, a phenomenon known as posterior collapse, which hind…
stat.ML2023
Controlling Continuous Relaxation for Combinatorial Optimization
Yuma Ichikawa
Unsupervised learning (UL)-based solvers for combinatorial optimization (CO) train a neural network that generates a soft solution by directly optimizing the CO objective using a c…
stat.ML2023★ 3 cited
High-dimensional Asymptotics of VAEs: Threshold of Posterior Collapse and Dataset-Size Dependence of Rate-Distortion Curve
Yuma Ichikawa, Koji Hukushima
In variational autoencoders (VAEs), the variational posterior often collapses to the prior, known as posterior collapse, which leads to poor representation learning quality. An adj…