Learnable multiclass hypothesis classes don't always have a sample compression scheme of fixed size.
problem The limitation of sample compression schemes for multiclass hypothesis classes.
method Analysis of DS dimension and sample compression schemes.
result Learnable multiclass hypothesis classes do not always have a sample compression scheme of fixed size.
Reduces multiclass and regression compression schemes to binary ones.
problem Developing efficient learning algorithms for multiclass and regression problems.
method Reduces sample compression schemes for binary classes to multiclass and regression settings.
result Establishes new compression schemes for multiclass and regression problems.
Positive results for agnostic regression with various losses.
problem Agnostic regression with bounded sample compression.
method Generic and efficient sample compression schemes for real-valued functions.
result Exact and approximate compression schemes for specific losses.
It was proved in 1998 by Ben-David and Litman that a concept space has a sample compression scheme of size d if and only if every finite subspace has a sample compression scheme of size d. In the compactness theorem, measurability of the hypotheses of the created sample compression scheme is not guaranteed; at the same…
BDC compresses both sample size and dimensionality of large datasets.
problem Large datasets in both sample size and dimensionality.
method Two-stage framework using Decoded MMD, Reconstruction MMD, and Encoded MMD.
result BDC achieves comparable or superior performance with lower cost and higher compression rates.
New method compresses large sample data for faster discriminant analysis.
problem Large sample sizes in discriminant analysis increase computational burden.
method Proposes a new compression approach for reducing training samples.
result Significant computational gains and superior predictive ability compared to random sub-sampling.
New bounds found for agnostic learning with sample compression schemes.
problem Finding optimal rates of convergence for agnostic learning.
method Established tight characterization of worst-case rates for agnostic learning with sample compression schemes.
result Optimal rates of convergence for size- k k k agnostic sample compression schemes are k log ( n / k ) n \sqrt{\frac{k \log(n/k)}{n}} n k l o g ( n / k ) . Reduces policy space complexity for reinforcement learning.
problem Efficiency in exploring vast policy spaces in reinforcement learning.
method Uses Rényi divergence and l 1 l_1 l 1 norm to determine sample size for accurate policy approximation. result Established error bounds for sample size requirements in model-based and model-free settings.
Adaptive sampling method optimizes DNN compression for resource-constrained platforms.
problem Efficiently compressing DNNs for resource-constrained platforms with high accuracy.
method Adaptive sampling using genetic algorithm-inspired operations to optimize hyperparameters.
result Adaptive sampling outperforms rule-based and reinforcement learning methods in compression rate and accuracy.
Clapping reduces memory usage in distributed optimization by reusing data samples.
problem Significant communication overhead and impractical memory overhead in pipeline-parallel distributed optimization.
method Lazy sampling strategy to reuse data samples across steps, supporting convergence without unbiased gradient assumptions.
result Clapping achieves convergence in few-epoch or online training regimes without sample-size memory overhead.
A new method for high-dimensional data classification reduces misclassification errors.
problem High-dimensional data classification with limited samples.
method Compressive Regularized Discriminant Analysis (CRDA) using joint-sparsity promoting hard thresholding and regularized covariance matrix estimators.
result CRDA gives fewer misclassification errors than competitors and accurately selects features.
New theory controls compression change probability without prior knowledge.
problem Controlling compression change probability without prior knowledge.
method New theory on compression function and statistical risk.
result Cardinality of compressed set is a consistent estimator of probability of change of compression.
Investigates principles of generalization in list learning, refutes sample compression conjecture.
problem Determining applicability of classical principles in list PAC learning.
method Examines uniform convergence and sample compression in list PAC learning.
result Sample compression fails in list PAC learning, refutes conjecture.
Efficiently compress neural networks by discarding redundant parameters.
problem Compressing neural networks to reduce computational and storage costs.
method Data-dependent coresets using importance sampling and sensitivity analysis.
result Proves the accuracy and generalization bounds of the compressed network.
A new framework for efficient large-scale learning using sketching of moments.
problem Efficiently learning from large datasets with limited computational resources.
method Compressing the training data into a low-dimensional sketch and solving a nonlinear least squares problem.
result Sufficient sketch sizes to control the generalization error of the procedure.
The article refines error bounds for various learning algorithms.
problem Achieving precise error rates for learning algorithms.
method General technique for obtaining bounds on error rates of sample-consistent classifiers.
result Refined bounds on error rates for several learning algorithms.
This paper automates deep model compression using reinforcement learning.
problem Efficiently compressing deep neural networks without sacrificing accuracy.
method Reinforcement learning-based actor-critic structure for automated compression.
result 4-fold reduction in FLOP with 2.8% higher accuracy for VGG-16.
Proposes a framework for private data augmentation in federated learning.
problem Privacy and performance issues in non-IID training datasets.
method Multi-hop federated augmentation with sample compression.
result Significantly improves privacy, transmission delay, and local training performance.
CTE improves explanation estimation with less data and faster computation.
problem Inefficient and inaccurate explanation estimation in machine learning models.
method Distribution compression through kernel thinning to reduce sample size.
result CTE significantly improves accuracy and stability of explanation estimation.
Compress++ speeds up distribution compression to near-linear time.
problem Accurately summarize a probability distribution using a small number of points efficiently.
method Introduces Compress++, a meta-procedure to speed up any thinning algorithm.
result Achieves n \sqrt{n} n points with O ( log n / n ) \mathcal{O}(\sqrt{\log n/n}) O ( log n / n ) integration error in O ( n log 3 n ) \mathcal{O}(n \log^3 n) O ( n log 3 n ) time and O ( n log 2 n ) \mathcal{O}( \sqrt{n} \log^2 n ) O ( n log 2 n ) space. Adaptive step-size method improves compressed SGD performance in machine learning.
problem Communication bottleneck in distributed and decentralized optimization.
method Developed an adaptive step-size method for compressed SGD.
result Order-optimal convergence rates for various objective functions.
A CAE improves DNN's outlier and adversary defense.
problem Improving DNN's robustness against outliers and adversaries.
method Proposes a classification-autoencoder (CAE) that compresses samples into disjoint spaces and uses a decoder to classify and defend against adversaries.
result The CAE achieves state-of-the-art outlier recognition and near-lossless classification of adversaries.
Study robust regression learning under adversarial attacks.
problem Understanding which function classes are learnable in the presence of adversarial attacks.
method Introduced a novel agnostic sample compression scheme and used fat-shattering dimension to construct adversarially robust sample compression schemes.
result Finite fat-shattering dimension classes are learnable in both realizable and agnostic settings.
Data-independent pruning method reduces neural network size with accuracy guarantees.
problem Limited computational and memory resources for neural networks.
method Structured pruning using coresets.
result First efficient algorithm with worst-case guarantees on compression and accuracy.
New method compresses neural networks using random code, improving efficiency.
problem Large memory footprint of deep neural networks.
method Training a variational distribution over weights, encoding using Kullback-Leibler divergence.
result Achieves state-of-the-art compression rates and test performance.
New study reveals how heavy-tailed SGD dynamics lead to compressible neural networks.
problem Understanding why large neural networks can be compressed effectively.
method Linking SGD dynamics to compressibility properties of neural networks.
result Large step-size/batch-size ratios and overparametrization lead to heavy-tailed SGD dynamics, making networks compressible.
Autoencoders achieve image compression without needing multiple transforms.
problem Learning transforms for image compression with varying quantization steps.
method Use a single learned transform for multiple rate-distortion points at test time.
result Comparable performance can be achieved with a single learned transform.
The paper bounds information losses in neural classifiers from sampling.
problem Information losses in neural classifiers from finite datasets.
method Proves a relationship between information losses and expected total variation of the estimated neural model, bounds this expected total variation as a function of dataset size.
result Obtains bounds on information losses that are less sensitive to input compression and much smaller than existing bounds.
Decomposable-Net compresses neural networks without retraining for various sizes.
problem Performance degradation when changing model size after training.
method Decomposes weight matrices via SVD and adjusts ranks for different sizes.
result Maintains and improves performance across multiple model sizes.
Boosting improves accuracy by combining weak learners into a voting classifier.
problem Boosting's theoretical performance is sub-optimal, especially for voting classifiers.
method Proposes a randomized boosting algorithm that outputs voting classifiers with a single logarithmic dependency on sample size.
result Randomized boosting achieves a generalization error with a single logarithmic dependency on the sample size.
DeepThin compresses deep neural networks, improving performance and reducing resource usage.
problem Efficiently compressing large neural networks for mobile devices.
method Combining rank factorization with a reshaping process to add nonlinearity.
result DeepThin achieves significant improvements in word error rates and test loss compared to existing methods.
We compress large neural networks for quick adaptation to specific contexts.
problem How to quickly adapt a pretrained large neural network to specific contexts.
method Propose a Bayesian hypernetwork framework to compress the network and encourage sparsity.
result Generated compressed networks are significantly smaller than baseline methods.
RFX accelerates and compresses Random Forests for large datasets.
problem Memory bottleneck in proximity matrices limits Random Forest analysis.
method QLORA compression, CPU TriBlock storage, GPU batch sizing, 3D MDS visualization.
result Proximity-based Random Forest analysis on larger datasets is feasible.
Compression method reduces word embedding size for NLP models.
problem Memory constraints in deploying deep learning models for NLP tasks.
method Low rank matrix factorization during training to compress word embeddings.
result 90% compression with minimal accuracy loss for sentence classification tasks.
We learn sparse precision matrices from compressed data sketches.
problem Learning a graph from high-dimensional data with limited storage.
method Estimate a sparse precision matrix from a sketch of the data using non-linear random features.
result It is possible to estimate a sparse precision matrix from a sketch of size $m=Ω\left((d+2k)\log(d)
ight)$ .
Data-independent neural pruning via coresets improves accuracy with 90% compression.
problem Efficiently reduce neural network size while maintaining accuracy.
method Data-independent neural pruning using coresets.
result 90% compression of LeNet-300-100 on MNIST with improved accuracy.
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.
This paper uses deep reinforcement learning to compress CNN models, reducing size and maintaining accuracy.
problem Reducing model size for efficient deployment on limited hardware resources.
method Two-stage compression pipeline: pruning and quantization using deep reinforcement learning.
result Significant reduction in model size with minimal loss in accuracy.
Paper proposes compressive ICA algorithms for ICA model.
problem Efficiently solving ICA model with reduced memory and computational complexity.
method Compressive learning approach to ICA model, proving existence of compressive ICA scheme, proposing two algorithms (IPG and ASD).
result Proposed algorithms achieve substantial memory gains over well-known ICA algorithms.
We propose a method for inferring the conditional indepen- dence graph (CIG) of a high-dimensional discrete-time Gaus- sian vector random process from finite-length observations. Our approach does not rely on a parametric model (such as, e.g., an autoregressive model) for the vector random process; rather, it only assu…
Condensa programmatically optimizes neural network compression.
problem Finding optimal compression strategies for neural networks.
method Bayesian optimization-based algorithm for automatic sparsity inference.
result Significant memory and runtime improvements for real-world DNNs.
We extend quantization-aware training to extreme model compression.
problem Maximizing model accuracy with minimal model size.
method Quantize a random subset of weights during training, allowing unbiased gradients through other weights.
result Established new state-of-the-art compromises between accuracy and model size.
IMPACT optimizes LLM compression by focusing on activation importance, reducing model size up to 55.4%.
problem Resource constraints in deploying large language models (LLMs).
method IMPACT integrates activation importance into low-rank compression, optimizing for both size and accuracy.
result IMPACT achieves up to 55.4% greater model size reduction while maintaining comparable or better accuracy.
A new embedding method for high-dimensional data.
problem Handling large sample sizes in high-dimensional spaces.
method Partitioning space into simplices and embedding into barycentric coordinates.
result Linear classifier in rich feature space yields highly non-linear decision boundaries.
Saec compresses recommendation system embeddings by clustering similar features.
problem Large embedding matrix in recommendation systems consumes excessive memory.
method Saec clusters similar features within a field to reduce embedding matrix size.
result Saec reduces embedding size by ~27x with no performance loss.
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.
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…
Paper optimizes privacy-preserving distribution estimation for sparse data.
problem Sparse distribution estimation under local differential privacy constraints.
method Compressive sensing approaches for privacy-preserving estimation.
result Significant reduction in sample complexity for approximately sparse distributions.