A new method, REC, compresses images by encoding their latent representations efficiently.
problem Efficiently compressing single images with latent representations.
method Relative Entropy Coding (REC) that directly encodes latent representations with codelength close to relative entropy.
result REC is more efficient for single image compression compared to previous methods and is competitive for lossy compression.
COIN++ compresses multiple data types efficiently.
problem Handling diverse data modalities in neural compression.
method Implicit neural representations and modulations quantization.
result Significant compression gains with reduced encoding time.
Compressed LLM embeddings improve noisy regression tasks without overfitting.
problem Noisy regression tasks with high signal-to-noise ratios.
method Comparison of embedding compression techniques using autoencoder hidden representations.
result Compression improves performance on noisy tasks like financial return prediction.
Surface parameterizations and registrations are important in computer graphics and imaging, where 1-1 correspondences between meshes are computed. In practice, surface maps are usually represented and stored as 3D coordinates each vertex is mapped to, which often requires lots of storage memory. This causes inconvenien…
Modality-agnostic compression improves across diverse data types.
problem Efficiently compressing data across multiple modalities.
method Functional view of data, Implicit Neural Representation (INR), modality-agnostic latent representations, variational compression.
result Improved performance compared to existing methods, especially for diverse modalities.
Compressed sensing techniques enable efficient acquisition and recovery of sparse, high-dimensional data signals via low-dimensional projections. In this work, we propose Uncertainty Autoencoders, a learning framework for unsupervised representation learning inspired by compressed sensing. We treat the low-dimensional …
HOPE uses Hilbert space to deconstruct deep network representations.
problem Deconstructing learned representations in deep networks is challenging.
method Introduces Hilbert Operator for Progressive Encoding (HOPE) to deconstruct network weights.
result HOPE provides an unbiased approach to network compression and fine-tuning.
This paper investigates compression techniques for deep neural networks to reduce their size without sacrificing performance.
problem Compression of large deep neural networks for resource-limited platforms.
method Weight pruning, quantization, and lossless weight matrix representations based on source coding.
result Achieved up to 165 times compression rate while maintaining or improving model performance.
This paper improves the scalability of sparse neural network compression.
problem Sparse neural network compression for diverse data modalities.
method State-of-the-art sparsification techniques and meta-learning.
result Meta-learning sparse compression networks achieve new state-of-the-art results.
Scattering representations simplify SBI for images without extra compression.
problem Efficiently performing simulation-based inference on images with limited data.
method Use scattering representations for compression and learning, combined with spatial averaging and expressive density estimators.
result Scattering representations provide more information than traditional methods, without requiring additional simulations.
The usage of deep generative models for image compression has led to impressive performance gains over classical codecs while neural video compression is still in its infancy. Here, we propose an end-to-end, deep generative modeling approach to compress temporal sequences with a focus on video. Our approach builds upon…
We describe a simple and general neural network weight compression approach, in which the network parameters (weights and biases) are represented in a "latent" space, amounting to a reparameterization. This space is equipped with a learned probability model, which is used to impose an entropy penalty on the parameter r…
Knowledge distillation (KD) is a popular method for reducing the computational overhead of deep network inference, in which the output of a teacher model is used to train a smaller, faster student model. Hint training (i.e., FitNets) extends KD by regressing a student model's intermediate representation to a teacher mo…
Let M be a hyperbolizable, nontrivial compression body without toroidal boundary components. In this paper, we characterize which discrete and faithful representations of the fundamental group of M into PSL(2,C) are separable-stable. The set of separable-stable representations forms a domain of discontinuity for the ac…
New method compresses facial videos using GANs and latent space optimization.
problem Efficiently compressing facial videos at low bit rates.
method Leverages StyleGAN for latent space representation and compression, learns optimal compression through entropy model and perceptual loss.
result Significantly reduces perceptual distortion at low bit rates compared to state-of-the-art codecs.
Introduces REVE, a regularization scheme that compresses class conditioned entropy.
problem Improving generalization performance of deep learning models.
method Identifies a variable responsible for final prediction, compresses class conditioned entropy, introduces a variational upper bound, and integrates a tractable loss into training.
result Demonstrates the efficiency of REVE on various neural networks and datasets.
New approach for sharing deep learning costs between devices and cloud.
problem Prohibitive deep learning computational requirements for embedded devices.
method Study of representation compressibility in MobileNetV2 for balancing computation, bandwidth, and accuracy.
result An optimal splitting layer for network can be found with a simple PCA-based compression scheme.
Paper establishes generalization bounds for representation learning using Minimum Description Length.
problem Designing efficient statistical supervised learning algorithms that generalize well to unseen data.
method Developed a compressibility framework using Minimum Description Length (MDL) to derive upper bounds on generalization error.
result Established the first theoretical generalization bounds for Information Bottleneck type encoders and representation learning.
This work analyzes how different forms of compressibility affect adversarial robustness in neural networks.
problem Understanding the interaction between compressibility and adversarial robustness in neural networks.
method Developed a principled framework to analyze the effects of neuron-level sparsity and spectral compressibility on adversarial robustness.
result Identified that different forms of compression can induce highly sensitive directions in the representation space that adversaries can exploit.
Proposes Decodable Information Bottleneck for optimal representation learning.
problem Finding optimal representations for supervised learning.
method Integrates information retention and compression with the desired predictive family.
result Optimal representations lead to better expected test performance and can be estimated with guarantees.
Paper tackles noisy labels by compressing feature representations.
problem Learning with noisy labels leads to overfitting and poor generalization.
method Introduces compression inductive bias using Dropout and Nested Dropout.
result Compression helps in combating label noise and improving performance.
A new method learns graph compression from data.
problem Graphs lack ordering, making conventional compression algorithms ineffective.
method Partition and Code framework: decompose, learn, encode.
result PnC achieves compression gains that grow with graph size.
We study the flow of information and the evolution of internal representations during deep neural network (DNN) training, aiming to demystify the compression aspect of the information bottleneck theory. The theory suggests that DNN training comprises a rapid fitting phase followed by a slower compression phase, in whic…
Paper presents J-RFDL for robust DL in compressed space, improving data representation robustness and accuracy.
problem Improving data representation robustness and accuracy in the presence of noise and outliers.
method Joint Robust Factorization and Projective Dictionary Learning (J-RFDL) in a factorized compressed space.
result Delivers superior performance in data representation and classification over state-of-the-art methods.
A new algorithm for compressing latent representations in deep models.
problem Compressing continuous latent representations in deep models.
method Separates model design and training from quantization; uses adaptive quantization based on posterior uncertainty.
result Image compression with the proposed algorithm outperforms JPEG over a wide range of bit rates.
Proposes a method to train neural networks directly on compressed text data.
problem Training neural networks on compressed text data without decompression.
method Introduces composer modules to encode symbols from grammar compression rules into vector representations.
result Demonstrates that the proposed method can achieve both memory and computational efficiency while maintaining moderate performance.
This work explores the non-convex optimization in compressive learning and the performance of heuristics.
problem The challenge of learning from compressed representations in compressive learning.
method Numerical simulations of the non-convex optimization landscape and heuristic performance.
result Properties of the non-convex optimization landscape and heuristic performance are explored.
Paper compresses RNNs using HT decomposition for better performance.
problem Large model sizes of RNNs in sequence analysis.
method Hierarchical Tucker (HT) tensor decomposition for model compression.
result HT-LSTM achieves better compression and accuracy than state-of-the-art methods.
Goal: This paper deals with the problems that some EEG signals have no good sparse representation and single channel processing is not computationally efficient in compressed sensing of multi-channel EEG signals. Methods: An optimization model with L0 norm and Schatten-0 norm is proposed to enforce cosparsity and low r…
This paper proposes a new method for efficient data compression using Bayesian neural networks.
problem Efficient compression of data represented as functions mapping coordinates to signal values.
method Overfitting variational Bayesian neural networks to the data and compressing an approximate posterior weight sample using relative entropy coding.
result Our method achieves strong performance on image and audio compression while retaining simplicity.
New findings show compressed representations are exponentially easier to learn.
problem Understanding the generalization power of neural networks.
method Studied the entropy of input variables as a simplicity assumption and proved a bound on sample complexity.
result Compressed representations are exponentially easier to learn, providing new insights into neural network generalization.
Principal component analysis, dictionary learning, and auto-encoders are all unsupervised methods for learning representations from a large amount of training data. In all these methods, the higher the dimensions of the input data, the longer it takes to learn. We introduce a class of neural networks, termed RandNet, f…
Model compression techniques, such as pruning and quantization, are becoming increasingly important to reduce the memory footprints and the amount of computations. Despite model size reduction, achieving performance enhancement on devices is, however, still challenging mainly due to the irregular representations of spa…
SANs use sparse activation functions to compress data representations.
problem Learning meaningful features without considering compression.
method Introduce φ metric, define activation functions, and present SANs.
result SANs achieve small description length and interpretable kernels.
In this work, we propose an end-to-end block-based auto-encoder system for image compression. We introduce novel contributions to neural-network based image compression, mainly in achieving binarization simulation, variable bit rates with multiple networks, entropy-friendly representations, inference-stage code optimiz…
Softmax temperature influences model representation rank and performance.
problem Understanding and optimizing softmax function's impact on model representations.
method Investigated softmax function's role in deep neural networks, introduced rank deficit bias.
result Softmax temperature affects model representation rank and can improve performance.
Improved neural image compression with refined latent representations.
problem Sub-optimal results from variational autoencoders due to imperfect optimization and capacity limitations.
method Stochastic Gumbel Annealing (SGA) and its extensions (SGA+), including three different methods.
result Significant improvement in compression performance, especially on the R-D trade-off.
Autoencoders compress and reconstruct data for various applications.
problem Efficiently compress and reconstruct data.
method Neural network architecture that encodes and decodes data.
result Autoencoders can be applied to various data types and applications.
New method improves image compression using bits-back coding.
problem Lossy image compression with deep latent variable models.
method Iterative inference, stochastic annealing, bits-back coding.
result New state-of-the-art performance on lossy image compression.
Study connects compressed signal to AWGN model for risk estimation.
problem Estimating high-dimensional signals under compression constraints.
method Utilizes Gaussian approximation and Wasserstein distance to relate compressed and noisy signals.
result Establishes a connection between estimator risks under different conditions.
The answers to many unsolved problems lie in the intractable chemical space of molecules and materials. Machine learning techniques are rapidly growing in popularity as a way to compress and explore chemical space efficiently. One of the most important aspects of machine learning techniques is representation through th…
Proposes a new method for clustering compressed data.
problem Limited communication bandwidth and low-power consumption.
method Joint Variational Autoencoders with Bernoulli mixture models (VAB).
result The model can perform clustering in the compressed data domain.
Paper proposes IIQ for compressing embedding vectors.
problem Memory issues in representing large vocabularies.
method Isotropic iterative quantization (IIQ) for binary compression.
result More than 30x compression ratio with comparable performance.
A new method uses a frozen language model to improve sample efficiency in reinforcement learning.
problem Improving sample efficiency in reinforcement learning with partially observable environments.
method FROZEN Hopfield network and HELM (History Embedding Language Model) method.
result HELM achieves new state-of-the-art results on Minigrid and Procgen environments.
The recent framework of compressive statistical learning aims at designing tractable learning algorithms that use only a heavily compressed representation-or sketch-of massive datasets. Compressive K-Means (CKM) is such a method: it estimates the centroids of data clusters from pooled, non-linear, random signatures of …
T-Basis represents neural network tensors with fewer parameters.
problem Efficiently representing neural network tensors with fewer parameters.
method T-Basis uses Tensor Rings to represent tensors in a neural network, parameterizing them with a small number of coefficients.
result T-Basis achieves high compression rates with minimal performance loss.
The topological information is essential for studying the relationship between nodes in a network. Recently, Network Representation Learning (NRL), which projects a network into a low-dimensional vector space, has been shown their advantages in analyzing large-scale networks. However, most existing NRL methods are desi…
Various forms of representations may arise in the many layers embedded in deep neural networks (DNNs). Of these, where can we find the most compact representation? We propose to use a pruning framework to answer this question: How compact can each layer be compressed, without losing performance? Most of the existing DN…