New sparsity operator reduces variance reduction methods' computational cost.
problem Reduce computational cost of variance reduction methods.
method Introduce random-top-k operator to estimate gradient sparsity and reduce operations per update.
result Our algorithm consistently outperforms SpiderBoost in various tasks.
Embeds sparsity in deep neural networks, allowing exact zero parameters.
problem Learning sparse structures in deep networks.
method Embeds sparsity into neural network structure, allowing exact zero parameters during training.
result Can learn both structured and unstructured sparsity.
Gradient descent implicitly favors group sparsity in neural networks.
problem Understanding implicit regularization in neural networks for structured sparsity.
method Novel neural reparameterization for diagonally grouped linear networks.
result Gradient descent without explicit regularization biases towards group sparsity.
Concerns about interpretability, computational resources, and principled inductive priors have motivated efforts to engineer sparse neural models for NLP tasks. If sparsity is important for NLP, might well-trained neural models naturally become roughly sparse? Using the Taxi-Euclidean norm to measure sparsity, we find …
A new method solves l1-regularized optimization problems efficiently and sparsely.
problem l1-regularized optimization problems in machine learning.
method Orthant Based Proximal Stochastic Gradient Method (OBProx-SG)
result Promotes sparsity of solutions substantially and converges to global optimal solutions.
A new method for differentiable structured sparsity improves neural network performance and sparsity.
problem Non-differentiability of structured sparsity penalties in neural networks.
method Introducing D-Gating, a differentiable approach to structured overparameterization. result The D-Gating objective converges to the L2,2/D-regularized loss and induces sparse learning dynamics. Gradient descent on normalized networks reveals sparsity preferences.
problem Understanding the inductive bias of gradient descent on normalized neural nets.
method Analysis of gradient descent on weight-normalized smooth homogeneous neural nets, focusing on SWN and EWN.
result EWN causes weights to be updated in a way that prefers asymptotic relative sparsity.
We consider the problem of learning a Gaussian variational approximation to the posterior distribution for a high-dimensional parameter, where we impose sparsity in the precision matrix to reflect appropriate conditional independence structure in the model. Incorporating sparsity in the precision matrix allows the Gaus…
Paper extends KPCA using dualization for faster, more robust algorithms.
problem Efficiently perform KPCA with robustness and sparsity.
method Dualization of convex functions for multiple objective functions, promoting sparsity and robustness.
result Significant speedup in KPCA training time and improved robustness and sparsity.
We investigate filter level sparsity that emerges in convolutional neural networks (CNNs) which employ Batch Normalization and ReLU activation, and are trained with adaptive gradient descent techniques and L2 regularization or weight decay. We conduct an extensive experimental study casting our initial findings into hy…
New algorithm reduces dimensionality in stochastic optimization.
problem Stochastic optimization in high-dimensional problems.
method Proposes a sparsity-inducing stochastic gradient-free (SI-SGF) algorithm.
result Proves dimension-free query complexity in convex and strongly convex cases.
DSA efficiently allocates sparsity across layers for budgeted pruning.
problem Efficiently distributing resources (sparsity) across layers in pruning under resource constraints.
method DSA uses differentiable pruning to find continuous layer-wise pruning ratios via gradient-based optimization.
result DSA achieves superior performance and significantly reduces the time cost of pruning.
A new algorithm estimates sparse gradients on graphs with improved risk bounds.
problem Estimating sparse gradients on graph-structured data.
method Tree-Projected Gradient Descent algorithm for gradient-sparse parameters.
result Achieves risk bound of ns∗log(1+s∗p). In seeking for sparse and efficient neural network models, many previous works investigated on enforcing L1 or L0 regularizers to encourage weight sparsity during training. The L0 regularizer measures the parameter sparsity directly and is invariant to the scaling of parameter values, but it cannot provide useful gradi…
Study the effects of data parallelism and sparsity on neural network training.
problem Understanding the effects of data parallelism and sparsity on neural network training.
method Conducted extensive experiments and developed a theoretical analysis.
result Found a general scaling trend between batch size and number of training steps to convergence for the effect of data parallelism, and difficulty of training under sparsity.
New DP optimization methods for sparse gradients, improving on existing algorithms.
problem Differentially private optimization with sparse gradients in high-dimensional settings.
method Improved bounds for mean estimation, pure- and approximate-DP algorithms for stochastic convex optimization.
result First nearly dimension-independent rates for DP optimization with sparse gradients.
Pruning neural networks improves feature learning in high-dimensional models.
problem Improving feature quality in high-dimensional statistical models.
method Demonstrated that pruning neural networks can optimally learn sparse models using gradient descent.
result Pruning neural networks proportional to the sparsity level of the model directions improves sample complexity.
The paper analyzes neural network dynamics after weights escape the origin.
problem Understanding gradient flow dynamics of neural networks after the origin.
method Analyzes gradient flow of homogeneous neural networks with locally Lipschitz gradients.
result Characterizes the first saddle point encountered after escaping the origin.
This paper improves federated learning efficiency by adaptively sparsifying gradients.
problem Efficiently training machine learning models with geographically dispersed data.
method Adaptive gradient sparsification for non-i.i.d. local datasets, fairness-aware, online learning approach.
result Up to 40% improvement in model accuracy for a finite training time.
Stochastic gradient descent (SGD) is commonly used for optimization in large-scale machine learning problems. Langford et al. (2009) introduce a sparse online learning method to induce sparsity via truncated gradient. With high-dimensional sparse data, however, the method suffers from slow convergence and high variance…
Exploiting sparsity enables hardware systems to run neural networks faster and more energy-efficiently. However, most prior sparsity-centric optimization techniques only accelerate the forward pass of neural networks and usually require an even longer training process with iterative pruning and retraining. We observe t…
Several convex formulation methods have been proposed previously for statistical estimation with structured sparsity as the prior. These methods often require a carefully tuned regularization parameter, often a cumbersome or heuristic exercise. Furthermore, the estimate that these methods produce might not belong to th…
New methods solve graph sparsity optimization problems faster.
problem Complex graph sparsity optimization problems in disease outbreak monitoring and social network analysis.
method Stochastic variance-reduced gradient-based methods GraphSVRG-IHT and GraphSCSG-IHT.
result Our methods achieve linear convergence speed.
The paper introduces a new method for feature selection without explicit sparsification.
problem Feature selection with implicit sparsity-inducing mechanisms.
method Optimization over a family of kernels with gradient descent.
result The method achieves exact sparsity without explicit penalization techniques.
Deep learning is finding its way into the embedded world with applications such as autonomous driving, smart sensors and aug- mented reality. However, the computation of deep neural networks is demanding in energy, compute power and memory. Various approaches have been investigated to reduce the necessary resources, on…
The mushroom body is the key network for the representation of learned olfactory stimuli in Drosophila and insects. The sparse activity of Kenyon cells, the principal neurons in the mushroom body, plays a key role in the learned classification of different odours. In the specific case of the fruit fly, the sparseness o…
Stochastic optimization algorithms update models with cheap per-iteration costs sequentially, which makes them amenable for large-scale data analysis. Such algorithms have been widely studied for structured sparse models where the sparsity information is very specific, e.g., convex sparsity-inducing norms or ℓ0-n…
Large-scale non-convex sparsity-constrained problems have recently gained extensive attention. Most existing deterministic optimization methods (e.g., GraSP) are not suitable for large-scale and high-dimensional problems, and thus stochastic optimization methods with hard thresholding (e.g., SVRGHT) become more attract…
AGS-CL selectively updates penalties based on node importance for continual learning.
problem Catastrophic forgetting in continual learning.
method Adaptive Group Sparsity (AGS) with proximal gradient descent.
result Significantly outperforms baselines on various continual learning benchmarks.
Hadamard Wirtinger Flow recovers sparse signals from fewer measurements.
problem Reconstructing sparse signals from magnitude-only measurements.
method Gradient descent with Hadamard parametrization (HWF).
result A single step of HWF recovers support from k(xmax∗)−2 samples. MoreauGrad interprets neural nets robustly and sparsely.
problem Lack of robustness and sparsity in gradient-based interpretation methods.
method Moreau envelope for smooth and robust interpretation, combined with L1 regularization for sparsity.
result MoreauGrad provides a smooth, robust, and sparse explanation of neural nets.
Autoencoders fail to capture sparse structure in 1-bit data compression.
problem Proving the performance of shallow autoencoders on sparse data compression.
method Gradient descent analysis and approximate message passing.
result Gradient descent minimizer for sparse data is the identity (up to permutation) above critical sparsity.
Tuning-free OR-PCA improves scalability for large datasets.
problem Dataset sensitivity of OR-PCA tuning parameters.
method Implicit regularization through modified gradient descents.
result Comparable or better performance on simulated and real-world datasets.
We propose two sparsity-aware normalized subband adaptive filter (NSAF) algorithms by using the gradient descent method to minimize a combination of the original NSAF cost function and the l1-norm penalty function on the filter coefficients. This l1-norm penalty exploits the sparsity of a system in the coefficients upd…
New sparsity attacks degrade DNN efficiency, raising concerns for resource-constrained systems.
problem Vulnerabilities in DNNs through energy and latency attacks.
method Proposed sparsity attacks that modify DNN inputs to reduce activation sparsity, increasing execution time and energy consumption.
result Adversarial sparsity attacks can degrade DNN efficiency by up to 1.82x in image recognition DNNs.
Hard Thresholding Pursuit (HTP) is an iterative greedy selection procedure for finding sparse solutions of underdetermined linear systems. This method has been shown to have strong theoretical guarantee and impressive numerical performance. In this paper, we generalize HTP from compressive sensing to a generic problem …
Our community has greatly improved the efficiency of deep learning applications, including by exploiting sparsity in inputs. Most of that work, though, is for inference, where weight sparsity is known statically, and/or for specialized hardware. We propose a scheme to leverage dynamic sparsity during training. In parti…
New insights explain speedup saturation in distributed learning with large batches and delays.
problem Understanding and optimizing speedup in distributed learning with large batches and delays.
method Theoretical analysis of strongly convex, convex, and non-convex settings, considering data sparsity.
result Identification of a data-dependent parameter explaining speedup saturation in both batch size and gradient staleness.
Powerpropagation makes neural networks inherently sparse.
problem Training sparse neural networks to reduce computational footprint and model size.
method Introduces a new weight-parameterisation technique exploiting gradient descent dynamics.
result Models trained with Powerpropagation have a higher density of zero weights, allowing for more efficient pruning.
Boosting as gradient descent algorithms is one popular method in machine learning. In this paper a novel Boosting-type algorithm is proposed based on restricted gradient descent with structural sparsity control whose underlying dynamics are governed by differential inclusions. In particular, we present an iterative reg…
SCOPE iteratively optimizes sparsity-constrained problems without tuning hyperparameters.
problem Optimizing sparsity-constrained problems in signal processing, statistics, and machine learning.
method SCOPE (Sparsity-Constrained Optimization via sPlicing itEration) replaces gradient steps with a splicing operation guided by the objective value.
result SCOPE achieves linear convergence and superior support recovery performance.
A new method reduces bias in adaptive Lasso estimates.
problem Bias in adaptive Lasso estimates.
method Proximal gradient approach to learn penalty coefficients as decision variables.
result Reduces bias in estimates and encourages arbitrary sparsity structure.
Improved kernel herding algorithm for faster quadrature rule convergence.
problem Slow convergence speed of standard kernel herding algorithm.
method Improved gradient approximation to obtain sparser solutions.
result The cosine of the angle between negative gradient and approximate gradient determines convergence speed.
SGD batch size affects autoencoder global minima sparsity and sharpness.
problem Investigating how batch size impacts autoencoder learning.
method Non-convex autoencoder training with SGD, varying batch sizes.
result SGD batch size influences global minimum sparsity and sharpness.
In recent works, both sparsity-based methods as well as learning-based methods have proven to be successful in solving several challenging linear inverse problems. However, sparsity priors for natural signals and images suffer from poor discriminative capability, while learning-based methods seldom provide concrete the…
Safe screening improves generalized CGM's feature selection stability.
problem Improving feature selection stability in generalized CGM.
method Coupling safe screening with generalized CGM.
result Safe screening matches solution support at rate O(1/(tδ2)). Proposes a boosting framework for sparsity in grouped covariates.
problem Sparsity and selection bias in grouped covariates.
method Component-wise and group-wise gradient boosting with adjusted degrees of freedom.
result Reduces bias and improves predictability in variable selection.
Paper proposes a new method to optimize deep neural networks with sparse regularization.
problem Difficulty in achieving optimal convergence rates for deep neural networks due to sparsity constraints.
method Introduces a novel penalized estimation method for sparse DNNs, resolving computational and theoretical issues.
result Establishes an oracle inequality for the excess risk of the proposed sparse-penalized DNN estimator and derives convergence rates.