Pith. sign in

Adaptive Discrete Communication Bottlenecks with Dynamic Vector Quantization

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
abstract

Vector Quantization (VQ) is a method for discretizing latent representations and has become a major part of the deep learning toolkit. It has been theoretically and empirically shown that discretization of representations leads to improved generalization, including in reinforcement learning where discretization can be used to bottleneck multi-agent communication to promote agent specialization and robustness. The discretization tightness of most VQ-based methods is defined by the number of discrete codes in the representation vector and the codebook size, which are fixed as hyperparameters. In this work, we propose learning to dynamically select discretization tightness conditioned on inputs, based on the hypothesis that data naturally contains variations in complexity that call for different levels of representational coarseness. We show that dynamically varying tightness in communication bottlenecks can improve model performance on visual reasoning and reinforcement learning tasks.

fields

cs.LG 1

years

2025 1

verdicts

REJECT 1

representative citing papers

Task-Driven Discrete Representation Learning

cs.LG · 2025-06-13 · reject · novelty 3.0

A task-conditioned discrete representation objective with a Wasserstein regularizer, claiming an accuracy versus sample-complexity trade-off, with applications to RL state abstraction and domain generalization.

citing papers explorer

Showing 1 of 1 citing paper.

  • Task-Driven Discrete Representation Learning cs.LG · 2025-06-13 · reject · none · ref 31 · internal anchor

    A task-conditioned discrete representation objective with a Wasserstein regularizer, claiming an accuracy versus sample-complexity trade-off, with applications to RL state abstraction and domain generalization.