Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,742 papers · 148 categories

Trend · papers per month

24477194 · Jun 202019922001200920172026
48 results for layerwise perturbation

New method μP2μP^2 improves neural network training by scaling perturbations layerwise.

problem Improving neural network performance as models scale up.
method Layerwise perturbation scaling in the infinite-width limit of neural networks.
result Layerwise perturbation scaling ensures all layers are effectively perturbed in the limit.

Use simplified layerwise linear models to understand neural dynamics.

problem Complex neural network dynamics are hard to grasp.
method Apply simplified layerwise linear models to explain neural phenomena.
result Simplified models explain neural collapse, emergence, etc.

A new method for learning Bayesian neural networks using layerwise inference.

problem Learning Bayesian neural networks efficiently and accurately.
method Bayesian layerwise inference, treating neural networks as stacked Bayesian linear models, with pseudo-targets defined by backpropagated gradients.
result The method converges quickly and performs well on various benchmarks.

This work proves the asymptotic freeness of layerwise Jacobians in MLPs with Haar orthogonal matrices.

problem Proving the asymptotic freeness of layerwise Jacobians in multilayer perceptrons (MLPs).
method Replacing each layer's parameter matrix with itself multiplied by a Haar orthogonal matrix, and using the invariance of the MLP.
result Proves the asymptotic freeness of layerwise Jacobians in MLPs with Haar orthogonal matrices.

Efficient algorithm for CMDPs reduces to offline density estimation.

problem Offline learning for CMDPs with horizon H.
method Reduction to offline density estimation, layerwise exploration-exploitation tradeoff.
result First efficient and near-optimal reduction from CMDPs to offline density estimation.

We improve neural network explainability by bypassing batch normalization.

problem Lack of transparency in neural networks.
method Layer-wise Relevance Propagation with a method to include normalization layers.
result Heatmaps are more accurate for convolutional layers with our method.

AlphaPruning optimizes LLM pruning using HT-SR theory for better performance.

problem Improving pruning of large language models to reduce size without sacrificing performance.
method AlphaPruning uses HT-SR theory to allocate layerwise sparsity ratios more theoretically.
result AlphaPruning prunes LLaMA-7B to 80% sparsity with reasonable perplexity.

Deep Gaussian Processes (DGP) are hierarchical generalizations of Gaussian Processes (GP) that have proven to work effectively on a multiple supervised regression tasks. They combine the well calibrated uncertainty estimates of GPs with the great flexibility of multilayer models. In DGPs, given the inputs, the outputs …

2018-01-09abs ↗pdf ↗

Estimating machine learning performance 'in the wild' is both an important and unsolved problem. In this paper, we seek to examine, understand, and predict the pointwise competence of classification models. Our contributions are twofold: First, we establish a statistically rigorous definition of competence that general…

2019-10-24abs ↗pdf ↗

SAEs struggle with curved activation manifolds, revealing layer-dependent scaling laws.

problem Sparse autoencoders' reconstruction error varies across layers, not fitting existing scaling laws.
method Cross-layer study of 844 SAE checkpoints, fitting and regressing on manifold geometry.
result Manifold geometry predicts layer-dependent width exponents in SAEs, with transferable coefficients.

New optimizers control network width scaling, improving stability and transfer across different model sizes.

problem Designing stable optimizers for networks of varying widths.
method Interpreting optimizers as steepest descent under mean-normalized operator norms, enabling layerwise composability and width-independent bounds.
result New optimizers like row normalization and column normalization provide stable learning-rate transfer across different model widths.

Proves deep networks can learn hierarchical structures efficiently.

problem Understanding how deep networks learn hierarchical structures in data.
method Random Hierarchy Models, gradient-based methods, layerwise training.
result Proves deep networks can efficiently learn hierarchical structures.

Injectivity of ReLU networks is characterized for generative models and inverse problems.

problem Injectivity in ReLU networks for generative models and inverse problems.
method Layerwise analysis, worst-case Lipschitz constants, differential topology, random projections.
result Global injectivity of ReLU networks requires expansivity between 3.4 and 10.5 for Gaussian matrices.

Develops methods to measure and set function-space learning rates in neural networks.

problem Measuring and optimizing changes in neural network output functions.
method Efficient methods to measure and set function-space learning rates, requiring minimal computational overhead.
result Demonstrates FLeRM (Function-space Learning Rate Matching) for hyperparameter transfer across model scales.

A new robust prefix-tuning framework improves model robustness against adversarial attacks.

problem Lack of robustness in prefix-tuning for adversarial attacks.
method Leveraging layerwise activations of pretrained models for additional prefix finetuning during the test phase.
result Framework substantially improves robustness over strong baselines while maintaining comparable accuracy on clean texts.

MTL-NAS combines NAS with GP-MTL for task-agnostic multi-task learning.

problem Designing architectures for diverse tasks with varying priors.
method Disentangled GP-MTL networks, hierarchical feature sharing, and gradient-based search.
result General-purpose model trained once can adapt to multiple tasks.

Deep neural networks (DNNs) depend on the storage of a large number of parameters, which consumes an important portion of the energy used during inference. This paper considers the case where the energy usage of memory elements can be reduced at the cost of reduced reliability. A training algorithm is proposed to optim…

2019-12-23abs ↗pdf ↗

Large learning rates work surprisingly well in standard parameterization, contrary to theory.

problem Theoretical limits of large learning rates do not match practical network behavior.
method Fine-grained analysis of learning rates and network behavior under cross-entropy loss.
result There are two distinct sub-regimes of unstable learning rates, with a controlled divergence regime where features continue to evolve.

We give algorithms with provable guarantees that learn a class of deep nets in the generative model view popularized by Hinton and others. Our generative model is an nn node multilayer neural net that has degree at most nγn^γ for some γ<1γ<1 and each edge has a random edge weight in [1,1][-1,1]. Our algorithm learns {\em …

2013-10-23abs ↗pdf ↗

The paper studies how noise synchronizes tokens in deep transformer models.

problem Understanding synchronization in deep learning models with noise.
method Proves convergence to a stochastic particle system and identifies the limiting SDE.
result The limiting model displays synchronization by noise and exponential dissipation of interaction energy.

Multitask learning has shown promising performance in many applications and many multitask models have been proposed. In order to identify an effective multitask model for a given multitask problem, we propose a learning framework called learning to multitask (L2MT). To achieve the goal, L2MT exploits historical multit…

2018-05-19abs ↗pdf ↗

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.

Gradient-based meta-learning methods leverage gradient descent to learn the commonalities among various tasks. While previous such methods have been successful in meta-learning tasks, they resort to simple gradient descent during meta-testing. Our primary contribution is the {\em MT-net}, which enables the meta-learner…

2018-01-17abs ↗pdf ↗

The paper tackles extrapolation of gene knockouts effects on RNA counts.

problem Modeling effects of gene knockouts on RNA counts for new perturbations.
method Formulated as a latent variable model with additive perturbation effects, proved identifiability, proposed PDAE for estimation.
result PDAE can accurately predict effects of unseen but identifiable perturbations.

TSSM splits neural networks for parallel training with minimal accuracy loss.

problem Accuracy degradation in parallel training of deep neural networks.
method TSSM reformulates alternating minimization to achieve parallelism with minimal accuracy loss.
result TSSM achieves significant speedup without accuracy loss on multiple datasets.

Novel geometry-informed irreversible perturbation accelerates Langevin dynamics convergence.

problem Accelerating convergence of Langevin dynamics for Bayesian computation.
method Geometry-informed irreversible perturbation of Riemannian manifold Langevin dynamics.
result Improves estimation performance over irreversible perturbations that ignore geometry.

Study linear perturbations in Schwarzschild black hole spacetime.

problem Linear perturbations of Schwarzschild black hole spacetime.
method Investigate linearised perturbation of constant mass aspect function foliation at null infinity.
result Linearised perturbations of Bondi energy and mass vanish, and all linear momentum can be achieved.

New research evaluates various perturbation methods for improving neural network robustness.

problem Understanding and improving robustness of Convolutional Neural Networks (CNNs) against adversarial attacks.
method Detailed evaluation of five main perturbation-based defenses, comparing random and deterministic approaches.
result Perturbation-based defenses are equivalent in efficacy, and attacks transfer between them.

We study the transfer of adversarial robustness of deep neural networks between different perturbation types. While most work on adversarial examples has focused on LL_\infty and L2L_2-bounded perturbations, these do not capture all types of perturbations available to an adversary. The present work evaluates 32 attack…

2019-05-03abs ↗pdf ↗

EVILL uses randomised perturbations to improve exploration in bandit problems.

problem Improving exploration in structured stochastic bandit problems.
method Solves for the minimiser of a linearly perturbed regularised negative log-likelihood function.
result EVILL matches the performance of Thompson-sampling-style methods in theory and practice.