HiLLoC compresses large images losslessly using VAEs.
problem Lossless compression of large color photographs.
method Fully convolutional VAE models trained on ImageNet are applied to lossless compression.
result Achieves state-of-the-art compression for full-size ImageNet images.
Develops a method for lossless compression using latent variable models.
problem Lossless compression of large datasets.
method Bits back with asymmetric numeral systems (BB-ANS) using latent variable models.
result Achieves state-of-the-art lossless compression of full-size colour images.
Bit-Swap improves lossless compression for hierarchical latent variable models.
problem Efficient lossless compression for latent variable models with hierarchical structure.
method Generalizes bits-back coding to hierarchical latent variable models with Markov chain structure.
result Achieves superior lossless compression rates for hierarchical latent variable models.
IDF uses integer flows for lossless compression of discrete data.
problem Lossless compression of discrete data with minimal reconstruction errors.
method Integer Discrete Flows (IDF) using integer maps and integer discrete coupling layers.
result IDFs achieve state-of-the-art lossless compression rates on various datasets.
IDF++ improves integer discrete flows for lossless compression.
problem Theoretical limitations of integer discrete flows for lossless compression.
method Investigated and improved integer discrete flows, addressing gradient bias and architecture modifications.
result Different architecture modifications improve integer discrete flows for lossless compression.
Study excess risk in statistical inference with transformations.
problem Excess risk in estimating random variables from feature vectors and transformations.
method Characterize lossless transformations, develop test statistics, and information-theoretic bounds.
result Strongly consistent partitioning test statistic for lossless transformations.
Maximizing margins leads to lossless compression of training data.
problem Generalization in supervised learning.
method Information-theoretic interpretation of margin maximization.
result Margin maximization is a form of lossless maximal compression.
This work improves DNN weight quantization with ADMM, achieving lossless binarization and reduced search space.
problem Improving DNN model compression and accuracy with low bit quantization.
method Extending ADMM framework for DNN weight quantization with progressive multi-step approach.
result Achieved lossless and fully binarized DNNs with reduced accuracy loss.
This study presents a new lossy image compression method that utilizes the multi-scale features of natural images. Our model consists of two networks: multi-scale lossy autoencoder and parallel multi-scale lossless coder. The multi-scale lossy autoencoder extracts the multi-scale image features to quantized variables a…
Lossless compression of deep neural networks using NTK and RMT.
problem Compressing large-scale deep neural networks for low-power devices.
method High-dimensional neural tangent kernel approach.
result Asymptotic spectral equivalence between NTK matrices of wide DNNs enables lossless compression.
Deep latent variable models have seen recent success in many data domains. Lossless compression is an application of these models which, despite having the potential to be highly useful, has yet to be implemented in a practical manner. We present `Bits Back with ANS' (BB-ANS), a scheme to perform lossless compression w…
Algorithm removes units and layers of neural networks without losing accuracy.
problem Deploying large neural networks under limited resources.
method Mixed-Integer Linear Programming (MILP) and L1 regularization.
result Lossless compression of neural networks is achieved.
A tutorial on VAEs explaining its derivation and applications.
problem Understanding the VAE model and its limitations.
method Explains VAE through probabilistic and information theoretic perspectives.
result Identifies two common misconceptions and their practical consequences.
NeLLoC improves image compression with parallel decoding.
problem Image compression with OOD generalization.
method Local autoregressive model with parallel decoding.
result Significant gains in compression runtime.
New method improves generalization of VAEs by reducing overfitting.
problem Generalization issues in VAEs overfitting to training data.
method Proposed a new training objective to improve amortized inference.
result Improved performance in image modeling and lossless compression.
DZip improves general-purpose lossless compression using neural networks.
problem Improving general-purpose lossless compression methods.
method Hybrid architecture combining adaptive and semi-adaptive training of neural networks for prediction followed by arithmetic coding.
result Significant compression improvements on various real and synthetic datasets.
Paper designs a lossy compression method for lossless prediction.
problem Ensuring high performance on predictive tasks with minimal data.
method Characterizes bit-rate requirements for invariant transformations, designs unsupervised objectives for neural compressors.
result Achieves substantial rate savings on ImageNet compared to JPEG without compromising classification performance.
Study reduces memory needs for active learning with enriched queries.
problem Expensive labeling costs in active learning.
method Introduces bounded memory active learning through enriched queries, introduces lossless sample compression.
result Can learn classifiers with bounded memory and query optimality.
ResRep prunes CNNs without losing accuracy by separating remembering and forgetting.
problem Pruning CNNs to reduce FLOPs without sacrificing accuracy.
method Decoupling remembering and forgetting in CNNs, using SGD for remembering and a novel update rule for forgetting.
result Achieved lossless pruning with high compression ratio (76.15% accuracy on ImageNet with 45% FLOPs reduction).
Statistical test detects model degradations in optimized language models.
problem Detecting model degradations in optimized language models.
method Statistical hypothesis testing framework based on McNemar's test.
result Even small accuracy degradations (0.3%) can be attributed to actual degradations, not noise.
SSFN self-estimates network size with low complexity and consistent performance.
problem Designing a self-estimating feed-forward network with low complexity and consistent performance.
method Joint optimization for layer and node estimation, low computational complexity, and use of lossless flow property and convex optimization.
result Consistent performance across Monte-Carlo trials and monotonically non-increasing cost with network growth.
Paper reduces vocabulary losslessly for language model cooperation.
problem Language models struggle to cooperate with different tokenizations.
method Established a theoretical framework for lossless vocabulary reduction.
result Efficiently converts models with different tokenizations to cooperate with maximal common vocabulary.
ARDMs are a new model class for autoregressive diffusion that generalize existing models and can compress data efficiently.
problem Efficient data compression and generation.
method Autoregressive Diffusion Models (ARDMs) that generalize existing autoregressive models and discrete diffusion models.
result ARDMs require significantly fewer steps for compression compared to discrete diffusion models.
Ensemble methods are among the state-of-the-art predictive modeling approaches. Applied to modern big data, these methods often require a large number of sub-learners, where the complexity of each learner typically grows with the size of the dataset. This phenomenon results in an increasing demand for storage space, wh…
The protection of user privacy is an important concern in machine learning, as evidenced by the rolling out of the General Data Protection Regulation (GDPR) in the European Union (EU) in May 2018. The GDPR is designed to give users more control over their personal data, which motivates us to explore machine learning fr…
Paper proposes a privacy-preserving method for estimating complex models.
problem Lack of flexibility in existing model classes for approximating data-generating processes.
method Privacy-preserving distributed estimation of generalized additive mixed models using component-wise gradient boosting.
result Proposed algorithm yields equivalent model estimates as component-wise gradient boosting on pooled data.
Generalizes bits back coding for time-series models with latent Markov structures.
problem Efficiently compressing time-series data with latent Markov structures.
method Extends bits back coding to time-series models with latent Markov structures, including HMMs and LGSSMs.
result Effective for small scale models, promising for larger scale settings like video compression.
The study proves Strichartz and spectral projection theorems on specific types of curved surfaces.
problem Proving Strichartz and spectral projection theorems on curved surfaces.
method Using large negative curvature neighborhoods, the study proves theorems on asymptotically conic and Euclidean ends surfaces.
result The study proves theorems without loss of interval on specific types of curved surfaces.
This book presents a methodology and philosophy of empirical science based on large scale lossless data compression. In this view a theory is scientific if it can be used to build a data compression program, and it is valuable if it can compress a standard benchmark database to a small size, taking into account the len…
Automatic recognition of the historical letters (XI-XVIII centuries) carved on the stoned walls of St.Sophia cathedral in Kyiv (Ukraine) was demonstrated by means of capsule deep learning neural network. It was applied to the image dataset of the carved Glagolitic and Cyrillic letters (CGCL), which was assembled and pr…
We introduce a new framework for unsupervised learning of representations based on a novel hierarchical decomposition of information. Intuitively, data is passed through a series of progressively fine-grained sieves. Each layer of the sieve recovers a single latent factor that is maximally informative about multivariat…
Stochastic variational inference (SVI) employs stochastic optimization to scale up Bayesian computation to massive data. Since SVI is at its core a stochastic gradient-based algorithm, horizontal parallelism can be harnessed to allow larger scale inference. We propose a lock-free parallel implementation for SVI which a…
We present a new similarity measure based on information theoretic measures which is superior than Normalized Compression Distance for clustering problems and inherits the useful properties of conditional Kolmogorov complexity. We show that Normalized Compression Dictionary Size and Normalized Compression Dictionary En…
Understanding the influence of features in machine learning is crucial to interpreting models and selecting the best features for classification. In this work we propose the use of principles from coalitional game theory to reason about importance of features. In particular, we propose the use of the Banzhaf power inde…
A method to analyze neural network performance by measuring layer saturation.
problem Understanding which layers contribute to network performance.
method Layer saturation method: restricts layer output to eigenspace of variance matrix.
result Layer saturation indicates which layers contribute to network performance.
New technique for flow models achieves theoretical compression lengths.
problem No guaranteed computationally efficient codes for flow models.
method Local bits-back coding for flow models.
result Efficient algorithms achieve theoretical codelengths for flow models.
Physics-inspired methods optimize SVD compression of LLMs.
problem Efficiently compressing large language models (LLMs) using SVD.
method FermiGrad for globally optimal rank selection and PivGa for lossless compression.
result Global optimization of SVD ranks and lossless compression of low-rank factors.
We present a scheme for online, unsupervised state discovery and detection from streaming, multi-featured, asynchronous data in high-frequency financial markets. Online feature correlations are computed using an unbiased, lossless Fourier estimator. A high-speed maximum likelihood clustering algorithm is then used to f…
PAQ8 is an open source lossless data compression algorithm that currently achieves the best compression rates on many benchmarks. This report presents a detailed description of PAQ8 from a statistical machine learning perspective. It shows that it is possible to understand some of the modules of PAQ8 and use this under…
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.
NeuZip compresses neural network weights to save memory during training and inference.
problem Memory constraints in neural network training and inference.
method Entropy-based dynamic weight compression.
result Significant reduction in memory usage without performance loss.
This work compresses sequences by treating them as continuous-time processes, enabling efficient discretization.
problem Efficient compression of sequences, especially with deep learning models that scale with sequence length.
method Treat sequences as continuous-time processes, learn efficient discretization, and decode at different time intervals.
result Automatic bit rate reductions in video and motion capture sequences using learned discretization.
Paper detects common subtrees with identical labels in trees.
problem Finding common subtrees with identical label distribution in tree data.
method Developed an algorithm for tree isomorphism and a new compression scheme for trees.
result The method efficiently finds and compresses common subtrees with identical labels.
Randomness and regularities in Finance are usually treated in probabilistic terms. In this paper, we develop a completely different approach in using a non-probabilistic framework based on the algorithmic information theory initially developed by Kolmogorov (1965). We present some elements of this theory and show why i…
Proposes a new linearity-based neural network compression method.
problem Reduction of neural network model size while maintaining accuracy.
method Integrates linearity-based intuition with ReLU activation functions to merge layers.
result Achieves up to 75% reduction in model size without loss of accuracy.
Neural NCD reveals LLMs don't compress well for classification.
problem The disconnect between compression and classification in neural networks.
method Developed Neural NCD to compare LLMs to classic algorithms, finding classification accuracy not correlated with compression rate.
result Classification accuracy is not predictable by compression rate alone, challenging current understanding.
Neural networks offer high-accuracy solutions to a range of problems, but are costly to run in production systems because of computational and memory requirements during a forward pass. Given a trained network, we propose a techique called Deep Learning Approximation to build a faster network in a tiny fraction of the …
Compressing neural nets is an active research problem, given the large size of state-of-the-art nets for tasks such as object recognition, and the computational limits imposed by mobile devices. We give a general formulation of model compression as constrained optimization. This includes many types of compression: quan…