Study on generalization in gradient-based meta-learning, showing flatter solutions and coherence between adaptation trajectories.
problem Understanding generalization in gradient-based meta-learning.
method Analysis of objective landscapes, experimental demonstration of solution properties, and empirical evidence on coherence between adaptation trajectories.
result Meta-test solutions become flatter, lower in loss, and further away from the meta-train solution as meta-training progresses, even as generalization starts to degrade.
Paper proposes a new method to measure model sensitivity using final model only.
problem Understanding model behavior using only the final trained model.
method Reframe TDA as measuring sensitivity, propose further training as gold standard, unify gradient-based methods.
result Gradient-based methods approximate further training but vary in quality.
Gradient-based method extracts slow features from high-dimensional data.
problem Extracting meaningful low-dimensional features from high-dimensional, temporally varying data.
method Power Slow Feature Analysis (PowerSFA) using gradient-based training of differentiable architectures.
result PowerSFA effectively extracts meaningful low-dimensional features in various data types.
Gradient-based training and pruning for radial basis function networks in materials physics.
problem Interpretable and robust machine learning for materials physics problems.
method Gradient-based training and pruning of radial basis function networks with closed-form optimization criteria.
result Pruned models provide compact and interpretable versions of larger models, offering insights into atom-level migration processes.
Fourier analysis explains why deep nets generalize well.
problem Understanding why deep nets generalize well despite having more parameters than data.
method Using Fourier analysis to study DNN training and generalization.
result DNNs prioritize low-frequency components during training and small initialization leads to good generalization.
Paper proposes HTAF for stable training of binary neural networks.
problem Challenges in training binary neural networks with gradient-based optimization.
method HTAF is a smooth approximation to the Heaviside function that enables stable training.
result HTAF enables stable training of various binary neural networks with gradient-based optimization.
Generative model initializes 2-layer network weights for small datasets.
problem Approximating functions with 2-layer networks using small datasets and gradient-based training.
method Initialize hidden weights with a learned proposal distribution parameterized as a deep generative model. Refine with gradient-based post-processing and regularization.
result Demonstrates effectiveness of the approach with numerical examples.
GFM models neural network training as a dynamical system to forecast final weights.
problem Computational intensity and inefficiency in training deep neural networks.
method Gradient Flow Matching (GFM) treats training as a dynamical system with learned vector fields.
result GFM achieves forecasting accuracy competitive with Transformer-based models and significantly outperforms classical baselines.
Paper presents an optimization-based attack and defense for graph neural networks.
problem Adversarial robustness of graph neural networks (GNNs).
method Gradient-based attack and optimization-based adversarial training.
result Optimization-based attack can significantly decrease GNN classification performance with minimal edge perturbations.
New gradient-based method learns causal structures from data.
problem Challenging task in causal structure learning.
method Graph Autoencoder framework for gradient-based optimization.
result Significantly outperforms other gradient-based methods on large causal graphs.
Bayesian Neural Networks are robust to gradient-based attacks in the large-data limit.
problem Vulnerability of deep learning models to adversarial attacks.
method Analysis of adversarial attacks in the large-data, overparametrized limit for Bayesian Neural Networks.
result BNN posteriors are robust to gradient-based adversarial attacks in the limit.
Study finds some datasets benefit from small training samples.
problem The effectiveness of large training datasets in modern computer vision.
method Gradient-based importance measure to analyze training images.
result Some datasets can be trained with smaller samples, others not.
Paper analyzes dataset distillation for efficient encoding of task-relevant information.
problem Efficiently encoding task-relevant information from gradient-based learning of non-linear tasks.
method Theoretical analysis of dataset distillation applied to two-layer neural networks with gradient-based training.
result Low-dimensional structure of the problem is efficiently encoded into distilled data, reproducing a model with high generalization ability.
New SMC method for pBNNs improves scalability and predictive performance.
problem Training pBNNs with high-dimensional stochastic parameters.
method Gradient-based proposals within SMC samplers.
result New method outperforms state-of-the-art in predictive performance and training time.
A new activation function k-WTA improves neural network defenses against adversarial attacks.
problem Improving neural network robustness against gradient-based adversarial attacks.
method Proposes k-Winners-Take-All activation function and analyzes its effectiveness.
result k-WTA activation significantly enhances neural network robustness against adversarial attacks.
Logit pairing methods don't provide robustness against adversarial examples.
problem Improving adversarial robustness of classifiers.
method Logit regularization methods (CLP, LSQ, ALP) to craft adversarial examples.
result Adversarial Logit Pairing (ALP) may provide robustness, but the increase is small.
Paper presents a novel gradient-based method for training models and hyperparameters simultaneously.
problem Achieving generalization in machine learning models.
method A novel gradient-based framework that trains parameters and hyperparameters simultaneously.
result Significantly smaller runtime compared to benchmark methods for equivalent prediction scores.
Algorithm optimizes millions of hyperparameters efficiently.
problem Training modern network architectures with millions of hyperparameters.
method Combines implicit function theorem with efficient inverse Hessian approximations for gradient-based optimization.
result Jointly tuning weights and hyperparameters is only a few times more costly than standard training.
A method to reduce Hessian matrix calculation cost in gradient-based meta-learning.
problem High memory footprint in calculating Hessian matrix for large-scale applications.
method Multi-step estimation of gradients to reuse the same gradient in a window of inner steps.
result Significant reduction in training time and memory usage with competitive or improved accuracies.
Neural networks perform differently when regression is treated as classification.
problem Understanding why neural networks perform better when regression is treated as classification.
method Analyzing two-layer ReLU networks and their feature spaces, focusing on the cross entropy loss vs. square loss.
result The support of the measure induced by the square loss differs from that of the cross entropy loss, indicating optimization difficulties.
Gradient-based methods can be biased by distributional asymmetries in bivariate categorical data.
problem Gradient-based causal discovery methods can be biased by distributional asymmetries in bivariate categorical data.
method Identified and examined two distributional biases: Marginal Distribution Asymmetry and Marginal Distribution Shift Asymmetry. Employed two simple models to demonstrate and control these biases.
result Gradient-based methods can be biased by distributional asymmetries, and these biases can be controlled.
New method reconstructs significant parts of training data from neural networks.
problem Understanding and reconstructing training data from neural networks.
method Proposes a novel reconstruction scheme based on recent theoretical results about neural network training.
result Shows that a significant fraction of training data can be reconstructed from neural network parameters.
A gradient-based method learns the structure of TAN for Bayesian network classifiers.
problem Learning the structure of Bayesian networks is difficult.
method A distribution over graph structures learned via gradient-based optimization.
result Consistently outperforms random and Chow-Liu TAN structures.
ASTRA improves TDA by more accurately approximating iHVP.
problem Improving insights into training data attribution.
method ASTRA uses EKFAC-preconditioner on Neumann series iterations to accurately approximate iHVP.
result Improving iHVP approximation significantly improves TDA performance.
MERL uses evolutionary and gradient-based methods to optimize sparse team-based and dense agent-specific rewards in multiagent coordination.
problem Training multiagent reinforcement learning policies on sparse team-based rewards is difficult and relying solely on agent-specific rewards is sub-optimal.
method MERL employs a split-level training platform with an evolutionary algorithm and a gradient-based optimizer, transferring skills between the two processes.
result MERL significantly outperforms state-of-the-art methods on coordination benchmarks.
This work bridges competitive learning with gradient-based learning for faster feature extraction.
problem Lack of powerful feature extractors in competitive learning methods.
method Introduces gradient-based competitive layers for feature extraction.
result Demonstrates theoretical equivalence and faster convergence of gradient-based competitive layers.
Gradient-based method prunes large models to create transferable architectures.
problem Creating transferable architectures from large models with limited fine-tuning data.
method Gradient-based algorithm for architecture pruning and subset selection.
result Successfully retrain architectures on new tasks with few fine-tuning data.
We examine the role of memorization in deep learning, drawing connections to capacity, generalization, and adversarial robustness. While deep networks are capable of memorizing noise data, our results suggest that they tend to prioritize learning simple patterns first. In our experiments, we expose qualitative differen…
Efficiently infers neural architectures for unseen tasks.
problem High cost of neural architecture search.
method Gradient-based framework sharing information across tasks.
result Quick identification of good candidate architectures for new tasks.
New algorithm estimates task affinities without repeated training, improving model performance and efficiency.
problem Efficiently estimating task affinities among multiple tasks for model training.
method Grad-TAG algorithm: trains a base model for all tasks and uses gradient-based linearization to estimate task affinities.
result Estimates task affinities with high accuracy and low computational cost.
We formulate a general framework for competitive gradient-based learning that encompasses a wide breadth of multi-agent learning algorithms, and analyze the limiting behavior of competitive gradient-based learning algorithms using dynamical systems theory. For both general-sum and potential games, we characterize a non…
This work investigates how gradient-based learning performs with structured data, revealing issues and improvements.
problem Gradient-based learning under structured data, particularly with a spiked covariance structure.
method Investigates the effect of a spiked covariance structure on gradient-based feature learning and proposes weight normalization.
result Gradient-based dynamics may fail to recover the true direction in anisotropic settings, but weight normalization can improve performance.
A novel gradient-based method optimizes decision trees for complex tasks.
problem Training decision trees with arbitrary differentiable loss functions.
method Gradient-based optimization using first and second derivatives of loss functions.
result Improves accuracy and flexibility in decision tree optimization.
This paper analyzes deep and wide transformer training dynamics.
problem Understanding the training dynamics of infinitely deep and wide transformers.
method Develops a mean-field framework for gradient-based training of transformers, controlling a neural PDE.
result Establishes a rigorous foundation for gradient-based transformer training, proving convergence to global minima.
Adaptor 'E' extends gradient-based optimizers to explore loss landscapes, improving generalization.
problem Finding lower and better-generalizing minima in deep learning.
method Proposes an adaptor 'E' to extend gradient-based optimizers, encouraging exploration along landscape valleys.
result Adapted optimizers increase test accuracy by an average of 2.5% in large-batch training tasks.
Differentiable pipeline replaces non-differentiable CAE components for shape optimization.
problem Gradient-based optimization is limited by non-differentiable components in CAE workflows.
method Surrogate models replace non-differentiable pipeline components, enabling gradient-based optimization.
result Gradient-based shape optimization possible without differentiable solvers.
New method handles uncertainty in adversarial attacks using ensemble noise simulation.
problem Uncertainty in adversarial attacks on neural networks.
method Simulates attacker's noisy perturbation using various gradient-based attack algorithms and a pre-processing Denoising Autoencoder (DAE) defense.
result Significant improvements in post-attack accuracy with the proposed ensemble-trained defense.
Paper proposes DSP to accelerate deep learning model training by addressing locking and straggler problems.
problem Training deep neural networks is slow and inefficient due to locking and straggler issues.
method Layer-wise Staleness and DSP algorithm to handle locking and straggler problems.
result DSP achieves significant training speedup with stronger robustness than compared methods.
New framework improves model robustness by focusing on stable relations across environments.
problem Standard supervised learning fails under data distribution shift.
method Gradient-based learning framework derived from the principle of independent causal mechanisms (ICM).
result Models generalize well to unseen scenarios, ignoring unstable relations.
Generalization in deep neural networks can be analyzed using minimax rates for gradient methods.
problem Generalization performance of over-parameterized neural networks
method Establishing a connection between gradient-based methods and kernel methods
result Deriving minimax-optimal rates for GD and SGD under polynomial network width scaling
Catastrophic forgetting is a problem faced by many machine learning models and algorithms. When trained on one task, then trained on a second task, many machine learning models "forget" how to perform the first task. This is widely believed to be a serious problem for neural networks. Here, we investigate the extent to…
Paper presents a new method for Bayesian deep learning that scales to Atari games.
problem Training neural networks on complex environments like Atari games is challenging.
method Adapted temporal difference Q-learning to work with Bayesian inference.
result TAGI allows for analytical inference of neural network parameters, achieving performance comparable to gradient-based methods.
Tricks adversarial attacks to target specific classes, improving classifier accuracy.
problem Recent adversarial defense approaches have failed to protect classifiers from untargeted attacks.
method Target Training defense tricks untargeted attacks into targeted attacks on designated classes, then derives the real class.
result 86.2% accuracy for CW-L2 (confidence=0) in CIFAR10, outperforming unsecured classifiers.
Combines gradient-based and competitive learning for unsupervised feature extraction.
problem Handling input data without supervision and replicating input manifold topology.
method Integrates gradient-based and competitive learning approaches to learn topological structures.
result The dual competitive layer outperforms the vanilla layer in high-dimensional datasets.
Linear algebra approach for parallel deep learning models.
problem Training large DNNs in distributed environments.
method Linear algebraic approach to model parallelism.
result Manual development of backward operators for gradient-based training.
Interval attacks find more adversarial examples than existing methods.
problem Evaluating robustness of adversarially trained neural networks against unknown attacks.
method Symbolic interval propagation for bound over-approximation and gradient-guided attacks.
result Interval attacks find on average 47% more violations than state-of-the-art methods.
Self-Tuning Networks optimize hyperparameters using bilevel optimization and gated best-response functions.
problem Optimizing hyperparameters for neural networks.
method Bilevel optimization with gated best-response functions to adapt regularization hyperparameters online.
result Self-Tuning Networks outperform fixed hyperparameter values on large-scale deep learning problems.
New scaling laws explain deep learning performance growth.
problem Understanding neural network performance growth.
method Analyzed entire training dynamics of various architectures.
result Identified two dynamical scaling laws.