Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userDeep Generative Models for Distribution-Preserving Lossy Compression
114
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
We propose and study the problem of distribution-preserving lossy compression. Motivated by recent advances in extreme image compression which allow to maintain artifact-free reconstructions even at very low bitrates, we propose to optimize the rate-distortion tradeoff under the constraint that the reconstructed samples follow the distribution of the training data. The resulting compression system recovers both ends of the spectrum: On one hand, at zero bitrate it learns a generative model of the data, and at high enough bitrates it achieves perfect reconstruction. Furthermore, for intermediate bitrates it smoothly interpolates between learning a generative model of the training data and perfectly reconstructing the training samples. We study several methods to approximately solve the proposed optimization problem, including a novel combination of Wasserstein GAN and Wasserstein Autoencoder, and present an extensive theoretical and empirical characterization of the proposed compression systems.
rate research
Read More
The field of deep generative modeling has succeeded in producing astonishingly realistic-seeming images and audio, but quantitative evaluation remains a challenge. Log-likelihood is an appealing metric due to its grounding in statistics and information theory, but it can be challenging to estimate for implicit generative models, and scalar-valued metrics give an incomplete picture of a models quality. In this work, we propose to use rate distortion (RD) curves to evaluate and compare deep generative models. While estimating RD curves is seemingly even more computationally demanding than log-likelihood estimation, we show that we can approximate the entire RD curve using nearly the same computations as were previously used to achieve a single log-likelihood estimate. We evaluate lossy compression rates of VAEs, GANs, and adversarial autoencoders (AAEs) on the MNIST and CIFAR10 datasets. Measuring the entire RD curve gives a more complete picture than scalar-valued metrics, and we arrive at a number of insights not obtainable from log-likelihoods alone.
When training end-to-end learned models for lossy compression, one has to balance the rate and distortion losses. This is typically done by manually setting a tradeoff parameter $beta$, an approach called $beta$-VAE. Using this approach it is difficult to target a specific rate or distortion value, because the result can be very sensitive to $beta$, and the appropriate value for $beta$ depends on the model and problem setup. As a result, model comparison requires extensive per-model $beta$-tuning, and producing a whole rate-distortion curve (by varying $beta$) for each model to be compared. We argue that the constrained optimization method of Rezende and Viola, 2018 is a lot more appropriate for training lossy compression models because it allows us to obtain the best possible rate subject to a distortion constraint. This enables pointwise model comparisons, by training two models with the same distortion target and comparing their rate. We show that the method does manage to satisfy the constraint on a realistic image compression task, outperforms a constrained optimization method based on a hinge-loss, and is more practical to use for model selection than a $beta$-VAE.
Many important data analysis applications present with severely imbalanced datasets with respect to the target variable. A typical example is medical image analysis, where positive samples are scarce, while performance is commonly estimated against the correct detection of these positive examples. We approach this challenge by formulating the problem as anomaly detection with generative models. We train a generative model without supervision on the `negative (common) datapoints and use this model to estimate the likelihood of unseen data. A successful model allows us to detect the `positive case as low likelihood datapoints. In this position paper, we present the use of state-of-the-art deep generative models (GAN and VAE) for the estimation of a likelihood of the data. Our results show that on the one hand both GANs and VAEs are able to separate the `positive and `negative samples in the MNIST case. On the other hand, for the NLST case, neither GANs nor VAEs were able to capture the complexity of the data and discriminate anomalies at the level that this task requires. These results show that even though there are a number of successes presented in the literature for using generative models in similar applications, there remain further challenges for broad successful implementation.
Most data is automatically collected and only ever seen by algorithms. Yet, data compressors preserve perceptual fidelity rather than just the information needed by algorithms performing downstream tasks. In this paper, we characterize the bit-rate required to ensure high performance on all predictive tasks that are invariant under a set of transformations, such as data augmentations. Based on our theory, we design unsupervised objectives for training neural compressors. Using these objectives, we train a generic image compressor that achieves substantial rate savings (more than $1000times$ on ImageNet) compared to JPEG on 8 datasets, without decreasing downstream classification performance.
In real-world applications, it is often expensive and time-consuming to obtain labeled examples. In such cases, knowledge transfer from related domains, where labels are abundant, could greatly reduce the need for extensive labeling efforts. In this scenario, transfer learning comes in hand. In this paper, we propose Deep Variational Transfer (DVT), a variational autoencoder that transfers knowledge across domains using a shared latent Gaussian mixture model. Thanks to the combination of a semi-supervised ELBO and parameters sharing across domains, we are able to simultaneously: (i) align all supervised examples of the same class into the same latent Gaussian Mixture component, independently from their domain; (ii) predict the class of unsupervised examples from different domains and use them to better model the occurring shifts. We perform tests on MNIST and USPS digits datasets, showing DVTs ability to perform transfer learning across heterogeneous datasets. Additionally, we present DVTs top classification performances on the MNIST semi-supervised learning challenge. We further validate DVT on a astronomical datasets. DVT achieves states-of-the-art classification performances, transferring knowledge across real stars surveys datasets, EROS, MACHO and HiTS, . In the worst performance, we double the achieved F1-score for rare classes. These experiments show DVTs ability to tackle all major challenges posed by transfer learning: different covariate distributions, different and highly imbalanced class distributions and different feature spaces.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
