New mechanisms from primate vision improve neural network robustness.
problem Demonstrating robust neural networks to small adversarial perturbations.
method Investigated two biologically plausible mechanisms: non-uniform retina sampling and receptive field diversity.
result Non-uniform retina sampling and receptive field diversity improve adversarial robustness.
Paper identifies how neural networks learn features.
problem Understanding how neural networks automatically select features.
method Deep Neural Feature Ansatz, implementing average gradient outer product.
result Recursive Feature Machines achieve state-of-the-art performance.
Statistical mechanics reveals phase transitions in ε-SVR error.
problem Understanding task precision in neural representations with variability.
method Statistical mechanics applied to ε-SVR. result Double-descent phenomenon in generalization error due to ε. Paper introduces a Gaussian Process for operator learning in computational mechanics.
problem Efficient and accurate solutions for large datasets with reliable uncertainty quantification.
method Gaussian Process (GP) embedded in a neural operator framework with stochastic dual descent (SDD) algorithm.
result Improves GP resolution independence and scalability for high-dimensional and non-linear systems.
This paper tackles the problem of defending a neural network against adversarial attacks crafted with different norms (in particular ℓ∞ and ℓ2 bounded adversarial examples). It has been observed that defense mechanisms designed to protect against one type of attacks often offer poor performance against…
Permutation-equivariant neural networks improve auction mechanisms by reducing regret and sample complexity.
problem Designing optimal auction mechanisms that balance revenue and bidders' regret.
method Introduced permutation-equivariant neural networks to auction mechanisms.
result Permutation-equivariant neural networks decrease expected ex-post regret and improve model generalizability.
Modern neural networks are often augmented with an attention mechanism, which tells the network where to focus within the input. We propose in this paper a new framework for sparse and structured attention, building upon a smoothed max operator. We show that the gradient of this operator defines a mapping from real val…
New mechanism protects neural network weights from privacy attacks during self-supervised learning.
problem Privacy risks during fine-tuning stage of self-supervised learning.
method Proposes a novel differential privacy mechanism using additive logistic noise.
result Reduces membership inference attack accuracy to 50% while maintaining below 5% performance loss.
In this paper, we focus on developing a novel mechanism to preserve differential privacy in deep neural networks, such that: (1) The privacy budget consumption is totally independent of the number of training steps; (2) It has the ability to adaptively inject noise into features based on the contribution of each to the…
New neural network improves audio classification accuracy.
problem Challenging audio classification problem in pattern recognition.
method Introduces a Classifier-Attention-Based Convolutional Neural Network (CAB-CNN) with an attention mechanism to reduce classifier complexity.
result Significantly improves audio classification performance, achieving more than 10% improvements.
Identifying computational mechanisms for memorization and retrieval of data is a long-standing problem at the intersection of machine learning and neuroscience. Our main finding is that standard overparameterized deep neural networks trained using standard optimization methods implement such a mechanism for real-valued…
Analyzes unsupervised neural networks using statistical mechanics and Monte Carlo simulations.
problem Understanding computational capabilities of unsupervised neural networks.
method Statistical mechanics approach and Monte Carlo simulations.
result Obtained a phase diagram summarizing network performance.
New statistical mechanics analysis shows edge pruning outperforms node pruning in neural networks.
problem Theoretical understanding of neural network pruning effectiveness is lacking.
method Statistical mechanics analysis of a teacher-student framework.
result DPP node pruning method is superior to other methods, but edge pruning is better overall.
DNNs generalize object recognition in novel orientations via neurons tuned to common features.
problem Understanding how DNNs generalize to objects in novel orientations.
method Training DNNs with familiar objects from multiple viewpoints and analyzing neuron responses.
result DNNs disseminate orientation-invariance from familiar objects to recognize objects in novel orientations.
A hybrid neural network improves robustness in estimating vehicle parameters from noisy data.
problem Estimating parameters of a mechanical vehicle model from noisy acceleration data.
method Introduced a convolutional neural network with two objective functions: naive and hybrid.
result The hybrid objective function outperforms the naive one in robustness on noisy input data.
Physics-informed neural networks improve model accuracy and efficiency.
problem Accurate dynamic models for technical systems are hard to achieve.
method Physics-informed neural ordinary differential equations (PINODE) integrating Lagrangian mechanics.
result Hybrid model combines physical insight and data approximation.
Noether's framework reveals symmetry-breaking in neural networks.
problem Understanding the role of symmetry breaking in neural networks.
method Developed a theoretical framework using Lagrangian mechanics.
result Identified 'kinetic symmetry breaking' and its effect on learning dynamics.
New attention mechanism improves meta-transfer learning in dynamic tasks.
problem Underfitting in meta-transfer learning with dynamic tasks.
method Proposed Recurrent Memory Reconstruction (RMR) attention mechanism.
result ASNP-RMR significantly outperforms baselines in various tasks.
This study approximates neural network features for modeling relations and attention mechanisms.
problem Approximating neural network features for modeling relations and attention mechanisms.
method Analyzes inner products of multi-layer perceptrons for universal approximation of symmetric and asymmetric relation functions.
result Universal approximation of relation functions and attention mechanisms using inner products of neural networks.
Neural Ordinary Differential Equation (Neural ODE) has been proposed as a continuous approximation to the ResNet architecture. Some commonly used regularization mechanisms in discrete neural networks (e.g. dropout, Gaussian noise) are missing in current Neural ODE networks. In this paper, we propose a new continuous ne…
Chemical transport models (CTMs), which simulate air pollution transport, transformation, and removal, are computationally expensive, largely because of the computational intensity of the chemical mechanisms: systems of coupled differential equations representing atmospheric chemistry. Here we investigate the potential…
New mechanism discovered for feature learning in CNNs.
problem Understanding how CNNs learn features from images.
method Proposed Convolutional Neural Feature Ansatz linking filter covariances to patch-based AGOPs.
result Deep ConvRFM algorithm learns features similar to deep CNNs, improving performance.
Efficient trainable front-end for neural speech enhancement.
problem Inefficient STFT front-ends in neural speech enhancement models.
method Butterfly mechanism for Fast Fourier Transform, trainable STFT window.
result Accuracy and efficiency improvements for low-compute systems.
AttViz offers online visualizations of neural language model attention mechanisms.
problem Limited interpretability of neural language models.
method Online toolkit for exploring self-attention mechanisms.
result Visualizations help understand model decision-making.
APD method decomposes neural network parameters into simple, faithful components.
problem Understanding the internal mechanisms learned by neural networks.
method Attribution-based Parameter Decomposition (APD) method.
result Demonstrated effectiveness in recovering features, separating computations, and identifying representations.
EML-CD discovers causal mechanisms from neural networks in a structured way.
problem Extracting causal mechanisms from neural network weights is ill-posed.
method Integrates EML operator into causal structure learning, representing each edge mechanism as a gated EML binary tree.
result Achieves SHD=11.2 +/- 0.4 on real data, matching or outperforming existing methods.
NNs accurately predict energy eigenvalues and other physical phenomena in 1D quantum mechanics.
problem Understanding how neural networks interpret physics.
method Training NNs to predict energy eigenvalues from potentials and testing their ability to generalize.
result NNs can predict physical phenomena not learned during training, indicating a new way of understanding physics.
We introduce a theory-driven mechanism for learning a neural network model that performs generative topology design in one shot given a problem setting, circumventing the conventional iterative process that computational design tasks usually entail. The proposed mechanism can lead to machines that quickly response to n…
Paper explains why plateau phenomenon is rare in modern deep learning.
problem Understanding why plateau phenomenon is less common in deep learning.
method Statistical mechanical analysis of neural network learning.
result Data with small and dispersed eigenvalues prevent plateau phenomenon.
In this work, we propose a generalized likelihood ratio method capable of training the artificial neural networks with some biological brain-like mechanisms,.e.g., (a) learning by the loss value, (b) learning via neurons with discontinuous activation and loss functions. The traditional back propagation method cannot tr…
A neural network learns from examples and optimizes by dreaming.
problem The gap between training data and biological neural networks' experience.
method Inspired by biological learning, a generalized Hopfield network with Hebbian learning and off-line sleeping mechanisms.
result The network learns from examples, generalizes, and optimizes its storage capacity.
As an important part of speech recognition technology, automatic speech keyword recognition has been intensively studied in recent years. Such technology becomes especially pivotal under situations with limited infrastructures and computational resources, such as voice command recognition in vehicles and robot interact…
Extracts important peaks from XRD spectra using Attention mechanism.
problem Identifying significant peaks in XRD patterns for material properties.
method Convolutional neural network with Attention mechanism to analyze deep features.
result Selected lattice constant predicts cathodic material cell voltage.
The paper discusses a pooling mechanism to induce subsampling in graph structured data and introduces it as a component of a graph convolutional neural network. The pooling mechanism builds on the Non-Negative Matrix Factorization (NMF) of a matrix representing node adjacency and node similarity as adaptively obtained …
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.
We propose multi-way, multilingual neural machine translation. The proposed approach enables a single neural translation model to translate between multiple languages, with a number of parameters that grows only linearly with the number of languages. This is made possible by having a single attention mechanism that is …
Study explores grokking in neural networks, revealing transition from memorization to generalization.
problem Understanding the transition from memorization to generalization in over-parameterized neural networks.
method Extensive experiments and exploration of various viewpoints on grokking mechanism.
result Sharp transition from no generalization to perfect generalization observed during prolonged training.
Approach to verify neural network training integrity.
problem Poisoning attacks during neural network training.
method Use of cryptographic mechanisms to verify training integrity.
result Provable verification of neural network training integrity.
Enhances neural networks in hyperbolic space for better data structure capture.
problem Capturing hierarchical data structures efficiently.
method Unified hyperbolic model for neural network components.
result Superior parameter efficiency and outperformance over Euclidean methods.
Neural HMM with AGA captures multi-scale dynamics in financial markets.
problem Capturing multi-scale temporal dynamics in financial markets.
method Parallel multi-resolution encoders, adaptive gating, and multi-head attention.
result Outperforms fixed-resolution baselines in predicting price movements and liquidity shocks.
Social-based recommendation systems exploit the selections of friends to combat the data sparsity on user preferences, and improve the recommendation accuracy of the collaborative filtering strategy. The main challenge is to capture and weigh friends' preferences, as in practice they do necessarily match. In this paper…
DIFNET tackles the suspended animation problem in deep graph neural networks.
problem Deep graph neural networks suffer from the suspended animation problem.
method DIFNET uses neural gates and graph residual learning for node hidden state modeling, and includes an attention mechanism for node neighborhood information diffusion.
result DIFNET effectively addresses the suspended animation problem and improves learning performance.
New methods predict language model out-of-distribution behaviors using causal mechanisms.
problem Predicting how language models behave on unseen data.
method Two methods: counterfactual simulation and value probing.
result Both methods achieve high AUC-ROC and outperform causal-agnostic approaches in out-of-distribution settings.
We describe an approach to understand the peculiar and counterintuitive generalization properties of deep neural networks. The approach involves going beyond worst-case theoretical capacity control frameworks that have been popular in machine learning in recent years to revisit old ideas in the statistical mechanics of…
Enhanced GNN with expanded attention window and partially random embeddings.
problem Limited expressivity of traditional GNNs in distinguishing non-isomorphic graphs.
method Graph attention network with expanding attention window and partially random initial embeddings. Head dropout for regularization.
result Improved ability to differentiate between non-isomorphic graphs.
EWS-GCN improves credit scoring by analyzing money transfer connections.
problem Improving credit scoring in transactional banking data.
method Edge Weight-Shared Graph Convolutional Network (EWS-GCN) combining graph and recurrent neural networks.
result EWS-GCN outperforms state-of-the-art models in credit scoring.
SANs estimate feature importance in neural models, identifying interactions and high-ranked features.
problem Understanding and interpreting black-box neural network models.
method Self-Attention Network (SAN) architecture for feature importance estimation.
result SANs identify similar high-ranked features and feature interactions as other methods, improving predictive performance.
Attention mechanisms in deep neural networks have achieved excellent performance on sequence-prediction tasks. Here, we show that these recently-proposed attention-based mechanisms---in particular, the Transformer with its parallelizable self-attention layers, and the Memory Fusion Network with attention across modalit…