Paper shows pre-training and transfer learning reduce sample complexity for neural networks.
problem Training high-dimensional supervised learning with limited labeled data.
method Study of single-layer neural networks via online stochastic gradient descent, considering concept shift.
result Pre-training and transfer learning reduce sample complexity by polynomial factors under general assumptions.
Training dynamics reveal neural network complexity.
problem Understanding the complexity of neural networks during training.
method Relating the Lipschitz constant and training behavior of neural networks.
result Steady training and bounded complexity in far regions of input space.
New framework explains how larger pre-trained models reduce downstream learning sample complexity.
problem Understanding why larger pre-trained models reduce sample complexity in downstream tasks.
method Introducing a novel framework called Caulking inspired by PEFT methods.
result Improved pre-trained models provably decrease downstream task sample complexity.
Cross-regularization adapts model complexity during training.
problem Manual tuning of model complexity for overfitting prevention.
method Directly adapts regularization parameters through validation gradients during training.
result Organic emergence of architecture-specific regularization during training.
New framework minimizes model complexity for improved few-shot learning.
problem Empirical benefits of pre-training scale with data size but lack theoretical explanation.
method Complexity Minimization framework for meta-representation learning.
result Theoretical analysis shows error rate improves with more meta-training data.
Paper discovers simplicial complexes connecting trained models for improved ensembling.
problem Improving robustness and accuracy of deep learning ensembles.
method Identifies mode-connecting simplicial complexes on loss surfaces.
result Efficiently builds simplicial complexes for ensembling, outperforming independent ensembles.
Study compares memorization of SimCLR to supervised and random labels training.
problem Understanding memorization in contrastive learning.
method Investigated SimCLR's memorization properties compared to supervised and random labels training.
result SimCLR's memorization is similar to random labels training in terms of training object complexity distribution.
Improved sample complexity for training diffusion models.
problem How many samples are needed to train an accurate diffusion model?
method Analyzing the sample complexity of training diffusion models using neural networks.
result Exponential improvement in the dependence on Wasserstein error and depth, along with improved dependencies on other parameters.
Adversarial training leads to clean data generalization with significant robust overfitting gap.
problem Significant robust generalization gap in adversarial training.
method Two theoretical views: representation complexity and training dynamics.
result ReLU nets with O(ND) extra parameters can achieve CGRO. Generative adversarial nets (GANs) and variational auto-encoders have significantly improved our distribution modeling capabilities, showing promise for dataset augmentation, image-to-image translation and feature learning. However, to model high-dimensional distributions, sequential training and stacked architectures …
New insights into training ReLU networks, especially as data dimensionality increases.
problem Understanding computational complexity of training ReLU networks with varying data dimensions.
method Analyzed the parameterized complexity of two-layer ReLU networks with respect to various loss functions, focusing on the influence of data dimensionality.
result Running time lower bounds and optimal brute-force strategies for training ReLU networks, extending previous results to broader loss functions.
New measure shows various training techniques control model complexity.
problem Understanding how to control model complexity in deep learning.
method Developed geometric complexity measure and demonstrated its effectiveness.
result Many training techniques control geometric complexity, providing a unified framework.
We show that the subsurface projection of a train track splitting sequence is an unparameterized quasi-geodesic in the curve complex of the subsurface. For the proof we introduce induced tracks, efficient position, and wide curves. This result is an important step in the proof that the disk complex is Gromov hyperbolic…
A new training method improves stability and generalization of DeepONets.
problem Training deep operator networks (DeepONets) is challenging due to nonconvex and nonlinear nature.
method Two-step training method: first train trunk network, then branch network. Introduced Gram-Schmidt orthonormalization.
result Generalization error estimate and numerical examples demonstrating effectiveness.
New bounds adaptively control spectral complexity of trained Transformers.
problem Understanding why Transformers generalize well in machine learning.
method Spectrum-adaptive post hoc generalization bounds for multi-layer Transformers.
result Bounds adaptively trade off spectral complexity against dimension and depth factors.
PINNs struggle with increasingly complex ODEs, especially when parameters control their complexity.
problem Evaluating physics-informed neural networks on complex coupled ODEs.
method Tuned benchmarks of partial differential equations and harmonic oscillators; varying network architecture and training method.
result PINNs fail to solve complex ODEs, revealing issues like insufficient capacity, poor conditioning, and high local curvature.
New method counts boundary pieces in ReLU classifiers for better complexity measure.
problem Current classification complexity measures are misleading and ineffective.
method Developed a novel method using tropical geometry to count exact boundary pieces.
result Boundary piece count is negatively correlated with robustness.
The high computational complexity associated with training deep neural networks limits online and real-time training on edge devices. This paper proposed an end-to-end training and inference scheme that eliminates multiplications by approximate operations in the log-domain which has the potential to significantly reduc…
Deep learning models, while effective and versatile, are becoming increasingly complex, often including multiple overlapping networks of arbitrary depths, multiple objectives and non-intuitive training methodologies. This makes it increasingly difficult for researchers and practitioners to design, train and understand …
Training neural networks is hard in fixed dimensions.
problem Training two-layer neural networks is computationally hard in fixed dimensions.
method Parameterized complexity analysis considering dimension and number of neurons.
result Training two-layer neural networks is NP-hard for two dimensions.
New algorithm trains deep neural networks without global optimization.
problem Training deep neural networks efficiently and without global optimization.
method Uses random complex exponential activation functions and Markov Chain Monte Carlo sampling.
result Consistently attains theoretical approximation rate for residual networks.
Proposes Neural Complexity (NC) for predicting and explaining generalization in deep neural networks.
problem Challenges in specifying a suitable complexity measure for deep neural networks to predict and explain generalization.
method A meta-learning framework that learns a scalar complexity measure through interactions with many heterogeneous tasks.
result Trained NC model can be added to standard training loss to regularize any task learner.
New algorithm trains neural nets on simple skills to learn complex tasks faster.
problem Learning complex tasks through simple imitation.
method Train neural networks on simple, easy-to-learn skills to accelerate learning of complex, hard-to-learn tasks.
result Consistently outperforms state-of-the-art baseline in training speed and performance.
Lower bounds on query complexity for reconstructing private learner's training data.
problem Query complexity of reconstructing private learner's training data.
method Minimax analysis, Rényi DP, Metric DP framework.
result First known lower bounds on adversary's query complexity for various DP learners.
SGD's performance improves with critical batch size, minimizing SFO complexity.
problem Optimizing SGD's performance with batch size and learning rate.
method Analysis of SGD using constant and decaying learning rates, focusing on batch size effects.
result SGD with critical batch size minimizes SFO complexity.
New method accelerates MBRL by 10x for complex quadrotor tasks.
problem Inefficient training of world models in MBRL for complex robotic tasks.
method Parallelized state-space models and privileged information for training.
result Significant speedup in world model and MBRL training times.
Proposes BN layers for neural networks on complex domains, improving training stability and accuracy.
problem Training stability and accuracy issues in neural networks on complex domains.
method Developed Riemannian batch normalization (BN) layers with connections to existing layers.
result Demonstrated improved performance on radar clutter classification, node classification, and action recognition.
Mixup reduces the sample complexity of finding optimal decision boundaries for more separable data.
problem Finding optimal decision boundaries in separable data distributions.
method Mixup technique applied to binary linear classification problems.
result Mixup significantly reduces the sample complexity for more separable data.
We analyze the computational complexity of Quantum Sparse Support Vector Machine, a linear classifier that minimizes the hinge loss and the L1 norm of the feature weights vector and relies on a quantum linear programming solver instead of a classical solver. Sparse SVM leads to sparse models that use only a small fr…
New federated learning framework reduces model complexity and improves performance.
problem Slow convergence in traditional federated learning due to non-i.i.d. data.
method Clients train personalized local models, server trains shared model, addressing heterogeneity.
result Substantial performance gains over baselines, robust to non-i.i.d. data.
Paper tightens lower bounds on decentralized training complexity.
problem Understanding and optimizing iteration complexity in decentralized training.
method Proved a tight lower bound on iteration complexity and proposed DeTAG algorithm.
result DeTAG achieves the theoretical lower bound with only a logarithmic gap.
Recent works have cast some light on the mystery of why deep nets fit any data and generalize despite being very overparametrized. This paper analyzes training and generalization for a simple 2-layer ReLU net with random initialization, and provides the following improvements over recent works: (i) Using a tighter char…
Persistently trained EBMs generate images and estimate complex densities.
problem Challenges in ML learning for energy-based models, especially non-convergence of MCMC.
method Introduce diffusion data, learn a joint EBM through persistent training with enhanced sampling.
result First simultaneous achievement of stability, post-training image generation, and superior out-of-distribution detection for image data.
We propose Rademacher complexity bounds for multiclass classifiers trained with a two-step semi-supervised model. In the first step, the algorithm partitions the partially labeled data and then identifies dense clusters containing κ predominant classes using the labeled training examples such that the proportion of t…
Neural nets learn simple distributions first, then more complex ones.
problem Understanding how neural networks generalize from simple to complex functions.
method Stochastic gradient descent training, synthetic data, CIFAR10, ImageNet pre-training.
result Neural networks initially use lower-order statistics, then higher-order ones.
Low complexity decentralized neural net with centralized performance.
problem Training large neural networks in distributed nodes without data sharing.
method Layer-wise learning using ADMM for low complexity and centralized performance.
result Equivalent learning performance to centralized training in distributed nodes.
Neurons perform computations, and convey the results of those computations through the statistical structure of their output spike trains. Here we present a practical method, grounded in the information-theoretic analysis of prediction, for inferring a minimal representation of that structure and for characterizing its…
This work simplifies Gaussian process regression for multiple outputs.
problem Exponential computational complexity in Gaussian process regression.
method Approximating the covariance kernel using eigenvalues and functions.
result Significant reduction in training and regression complexity.
DIVA generates diverse tasks for complex simulators, enabling adaptive agent training.
problem Lack of diverse training data for complex, open-ended simulators.
method Evolutionary approach using domain randomization and procedural generation.
result Successfully trains adaptive agent behavior in complex simulators.
Optimal batch size minimizes training time for neural networks.
problem Minimizing training time for two-layer neural networks with SGD.
method Characterized optimal batch size as a function of target hardness (information exponents). Used Correlation loss SGD to overcome limitations.
result Optimal batch size minimizes training time without changing total sample complexity.
Pre-training is crucial for learning deep neural networks. Most of existing pre-training methods train simple models (e.g., restricted Boltzmann machines) and then stack them layer by layer to form the deep structure. This layer-wise pre-training has found strong theoretical foundation and broad empirical support. Howe…
The high computational and parameter complexity of neural networks makes their training very slow and difficult to deploy on energy and storage-constrained computing systems. Many network complexity reduction techniques have been proposed including fixed-point implementation. However, a systematic approach for designin…
The FloatSD technology has been shown to have excellent performance on low-complexity convolutional neural networks (CNNs) training and inference. In this paper, we applied FloatSD to recurrent neural networks (RNNs), specifically long short-term memory (LSTM). In addition to FloatSD weight representation, we quantized…
A persistent challenge in practical classification tasks is that labeled training sets are not always available. In particle physics, this challenge is surmounted by the use of simulations. These simulations accurately reproduce most features of data, but cannot be trusted to capture all of the complex correlations exp…
Deep reinforcement learning approaches have shown impressive results in a variety of different domains, however, more complex heterogeneous architectures such as world models require the different neural components to be trained separately instead of end-to-end. While a simple genetic algorithm recently showed end-to-e…
Discrete random variables are natural components of probabilistic clustering models. A number of VAE variants with discrete latent variables have been developed. Training such methods requires marginalizing over the discrete latent variables, causing training time complexity to be linear in the number clusters. By appl…
Changing kernel bandwidth during training improves kernel regression performance.
problem Improving kernel regression performance with varying model complexity.
method Investigated changing the bandwidth of a translational-invariant kernel during training for kernel regression using gradient descent.
result Kernel regression exhibits double descent behavior with decreasing model complexity (bandwidth).
Study the expressivity and training complexity of polynomial neural networks.
problem Understanding the expressivity and training complexity of polynomial neural networks.
method Use algebraic geometry to describe neuromanifolds and neurovarieties, analyzing their dimension and learning degree.
result Characterized the dimension and learning degree of neuromanifolds, providing geometric and complexity measures.