Reducing the Computational Cost of Deep Generative Models with Binary Neural Networks
arXiv:2010.13476
Abstract
Deep generative models provide a powerful set of tools to understand real-world data. But as these models improve, they increase in size and complexity, so their computational cost in memory and execution time grows. Using binary weights in neural networks is one method which has shown promise in reducing this cost. However, whether binary neural networks can be used in generative models is an open problem. In this work we show, for the first time, that we can successfully train generative models which utilize binary neural networks. This reduces the computational cost of the models massively. We develop a new class of binary weight normalization, and provide insights for architecture designs of these binarized generative models. We demonstrate that two state-of-the-art deep generative models, the ResNet VAE and Flow++ models, can be binarized effectively using these techniques. We train binary models that achieve loss values close to those of the regular models but are 90%-94% smaller in size, and also allow significant speed-ups in execution time.
Accepted to ICLR 2021
References in corpus (8)
- NICE: Non-linear Independent Components Estimation
- Flow++: Improving Flow-Based Generative Models with Variational Dequantization and Architecture Design
- BIVA: A Very Deep Hierarchy of Latent Variables for Generative Modeling
- Bit-Swap: Recursive Bits-Back Coding for Lossless Compression with Hierarchical Latent Variables
- Practical Lossless Compression with Latent Variables using Bits Back Coding
- Compression with Flows via Local Bits-Back Coding
- Projection Convolutional Neural Networks for 1-bit CNNs via Discrete Back Propagation
- Espresso: Efficient Forward Propagation for BCNNs