Dynamic steerable blocks improve deep networks by learning filter invariances.
problem Pixel-based filters ignore image properties, leading to suboptimal performance.
method Developed frame-based ResNets and Densenets, which are steerable under predefined transformations.
result Dynamic steerable blocks outperform other approaches on contour detection datasets.
Steerable CNNs boost image classification accuracy.
problem Improving accuracy in image classification tasks.
method Presented an efficient, flexible class of equivariant convolutional networks.
result Achieved state-of-the-art results on CIFAR image classification benchmark.
Scale-equivariant CNNs handle scale changes for improved performance.
problem Translation equivariance is not sufficient for handling scale changes in CNNs.
method Developed scale-equivariant convolutional networks with steerable filters.
result Demonstrated state-of-the-art results on MNIST-scale and STL-10 datasets.
Steerable E(3) Graph Neural Networks incorporate geometric and physical covariant information.
problem Incorporating covariant information like position, force, velocity, or spin in graph neural networks.
method Steerable E(3) Equivariant Graph Neural Networks (SEGNNs) that use steerable MLPs to incorporate geometric and physical covariant information.
result SEGNNs improve upon classic linear point convolutions and recent equivariant graph networks that send invariant messages.
Steerable neural ODEs on homogeneous spaces for equivariant feature dynamics.
problem Learning continuous-time equivariant dynamics of vector-valued features on homogeneous spaces.
method Introduces steerable neural ordinary differential equations on homogeneous spaces, interpreting features as sections of associated vector bundles over M. result Steerable NODEs are G-equivariant when the flow and connection are G-invariant, and they incorporate existing models. 3D Steerable CNNs learn equivariant features for 3D data.
problem Learning rotationally equivariant features in volumetric data.
method SE(3)-equivariant convolutions using steerable kernel basis.
result 3D Steerable CNNs are effective for protein structure classification and amino acid propensity prediction.
RotDCF decomposes CNN filters for rotation-equivariant deep networks.
problem Handling global deformations in images for vision tasks.
method Decomposes convolutional filters over joint steerable bases for rotation-equivariance.
result Significantly reduces model size and computational complexity while preserving performance.
New method identifies how platforms can influence consumer behavior.
problem Estimating the causal effect of digital platforms on consumption.
method General causal inference problem, focusing on observational designs, and explicitly modeling consumption dynamics.
result Exogenous variation in consumption and responsive algorithmic control actions are sufficient for identifying steerability of consumption.
The paper generalizes equivariant neural networks on homogeneous spaces to the non-linear setting.
problem Equivariant neural networks on homogeneous spaces.
method Deriving generalized steerability constraints for non-linear equivariant layers.
result The universality of the derived construction for non-linear equivariant layers.
Geometric stability predicts steerability and detects drift in language models.
problem Predicting steerability and detecting drift in language models.
method Supervised and unsupervised geometric stability measures.
result Supervised geometric stability predicts steerability with high accuracy and detects drift earlier.
A new model for complex cells accounts for insensitivity to image shifts.
problem Complex cells' response to image shifts.
method Linear combination of Gaussian derivatives at a single position.
result Maximum response insensitive to small shifts of the image.
This paper establishes a mathematical framework for G-CNNs on homogeneous spaces.
problem Designing equivariant neural networks for data with symmetries.
method Using Mackey's theory on induced representations, the paper presents a general framework for G-CNNs.
result G-CNNs are a universal class of equivariant network architectures.
New model improves field learning with improved equivariance.
problem Learning equivariant stochastic fields.
method Equivariant Gaussian processes and Steerable Conditional Neural Processes.
result SteerCNPs significantly improve performance in transfer learning tasks.
3D scattering model predicts lithium-silicon formation energies.
problem Predicting formation energies of amorphous Li-Si materials.
method Steerable wavelet scattering for 3D signals, invariant to translations and rotations.
result State-of-the-art results compared to other machine learning methods.
ProSeNet provides interpretable deep sequence models with natural explanations.
problem Challenges in explaining deep neural network predictions for sequence modeling.
method Prototypes derived from case-based reasoning, with criteria for simplicity, diversity, and sparsity.
result Achieves accuracy on par with state-of-the-art models while providing interpretable explanations.
GSA-Nets apply group equivariance to self-attention for vision tasks.
problem Improving self-attention networks for vision tasks.
method Define group-equivariant positional encodings.
result GSA-Nets outperform non-equivariant self-attention networks on vision benchmarks.
Regularization improves stability and consistency of sparse autoencoders.
problem Varying features across random seeds and training choices in SAEs.
method Added L1 or L2 penalties on encoder and decoder weights.
result L2 regularization increases cross-seed feature consistency.
Paper proves spectral filters can be transferred between graphs.
problem Proving spectral filters can be transferred between graphs.
method Introducing the Cayley smoothness space and proving filters in this space are linearly stable.
result Graph spectral filters are transferable if they are in the Cayley smoothness space.
This work prunes CNN filters based on their functionality, not just size.
problem Redundant filters in CNNs waste computation resources.
method Functionality-oriented filter pruning method.
result Pruning based on functionality optimizes computation and interprets filter importance.
A new SOHP filter improves trend estimation in economic time series.
problem Improving trend estimation in nonlinear economic time series.
method Recursive application of one-sided HP filter on updated cyclical components, combined with an incremental HP filtering algorithm.
result Better performance of SOHP filter compared to other HP-type filters on real economic data.
Deep density methods improve filtering in high-dimensional systems.
problem Nonlinear filtering in high-dimensional systems.
method Two deep density methods based on Feynman-Kac formulas and neural networks.
result Logarithmic deep backward stochastic differential equation filter outperforms classical methods in high dimensions.
The paper explores modifications to filter banks for speech recognition.
problem Improving speech recognition accuracy using modified filter banks.
method The authors investigate replacing triangular filters with Gabor or Gammatone filters, and rearranging filter bank computations to integrate features over smaller time scales.
result No significant improvements in phone error rate were observed with the modifications.
Pruning filters in CNNs improves interpretability, showing shape-selective filters are crucial for object recognition.
problem Interpreting the complex decision-making process of CNNs is challenging due to their large number of parameters.
method We developed a greedy structural compression scheme that prunes filters based on the classification accuracy reduction (CAR) index.
result Pruned filters in CNNs, especially those in the first and second layers, are more likely to be shape-selective, indicating their importance in object recognition.
Gradient filters track moving parameters under noisy data and misspecification.
problem Tracking multidimensional time-varying parameters under noisy observations and model misspecification.
method Gradient-based filters update parameters using the gradient of a postulated objective function, evaluated at either the predicted or updated parameters.
result Novel sufficient conditions for exponential stability of the filtered parameter path, and finite-sample and asymptotic mean squared error bounds.
We simplify Bayesian filtering by framing it as optimization, making it practical for high-dimensional systems.
problem Bayesian filtering struggles in high-dimensional state spaces like neural networks.
method We frame Bayesian filtering as optimization, using gradient descent for nonlinear cases.
result Our method results in effective, robust, and scalable filters for high-dimensional systems.
A new model optimizes Bloom filters using machine learning.
problem Improving the efficiency of Bloom filters for data sets.
method Modeling learned Bloom filters with machine learning, optimizing with sandwiching method.
result Optimized learned Bloom filters provide improved performance.
Group equivariant neural networks simplify complex tasks with group representation theory.
problem Challenging tasks requiring input transformations like rotations.
method Group representation theory, non-commutative harmonic analysis, differential geometry.
result A neural network is group equivariant if and only if it has a convolutional structure.
Develops an inverse particle filter for cognitive systems.
problem Tracking cognitive adversaries in counter-adversarial applications.
method Global filtering approach using Monte Carlo methods and differentiable I-PF.
result Demonstrates convergence to optimal inverse filter and improved estimation performance.
A novel method reduces dimensionality for filtering SRNs with observed variables.
problem Challenges in estimating hidden state variables in SRNs with limited observations.
method Filtered Markovian Projection (Filtered MP) for dimensionality reduction in filtering.
result Filtered MP guarantees consistency and superior computational efficiency in high dimensions.
BlosSOM improves data visualization for large datasets.
problem Insufficient performance of dimensionality reduction methods for large datasets.
method GPU-accelerated semi-supervised EmbedSOM algorithm.
result Produces high-quality visualizations with user control.
Kernel learning FBSDE filter improves nonlinear filtering efficiency.
problem Nonlinear filtering problem in high-dimensional systems.
method Iterative and adaptive meshfree approach using forward backward SDE and KDE.
result Rigorous convergence analysis provided, supporting empirical results.
Paper develops a particle filter for rapid model parameter adaptation and change detection.
problem Rapidly adapting to changes in model parameters and distinguishing between regime shifts and stochastic volatility.
method Incorporates genetic algorithm elements into a particle filter for accelerated adaptation and change detection.
result The filter adapts to regime shifts extremely rapidly and provides a clear heuristic for distinguishing between regime shifts and stochastic volatility.
New method filters large networks from financial data to reveal key subnetworks.
problem Filtering large dimensional networks to isolate key constituents.
method Exploits spectral properties of high-dimensional data networks, tuning for sparsity and consistency.
result Shows method can interpolate between zero and maximal filtering, preserving spectral properties.
Many nonlinear extensions of the Kalman filter, e.g., the extended and the unscented Kalman filter, reduce the state densities to Gaussian densities. This approximation gives sufficient results in many cases. However, this filters only estimate states that are correlated with the observation. Therefore, sequential esti…
Constructs tangent groupoid for filtered manifolds without coordinates.
problem No specific problem stated; intrinsic construction of tangent groupoid.
method Intrinsic construction of tangent groupoid.
result Intrinsic construction of tangent groupoid for filtered manifolds.
This work analyzes the stability of graph filters under large perturbations.
problem Stability of graph filters under large edge rewires.
method Proves a bound on stability using frequency response and community structure.
result Graph filter stability depends on perturbation to community structure.
Improved Kalman filter for non-linear, non-Gaussian data.
problem Estimating hidden variables with non-linear, non-Gaussian observations.
method Reproduces and extends Burkhart et al.'s discriminative Kalman filter.
result Enhanced filter performance for complex observation models.
Paper proves convergence of Kalman filter on Stiefel manifolds with measurement errors.
problem Filtering constant particle with measurement errors on Stiefel manifolds.
method Extended Kalman filter applied to Stiefel manifold-valued observations.
result Convergence of the extended Kalman filter proved for constant system process.
Advances deep network embedding through multi-filtering GCN.
problem Existing attribute embedding methods fail to capture different aspects of node features.
method Multi-filtering Graph Convolution Neural Network (GCN) framework.
result Significant improvement in link prediction and node classification tasks with limited training data.
Net2Vec maps filters to vectors to reveal complex concept encoding.
problem Understanding how deep neural networks encode semantic concepts.
method Net2Vec framework that maps semantic concepts to vectorial embeddings based on filter responses.
result Multiple filters are often required to code for a concept, and filters help encode multiple concepts.
This work preserves linear invariants in ensemble filters for non-Gaussian data assimilation.
problem Maintaining critical invariants like mass, stoichiometric balance, and charge in non-Gaussian data assimilation.
method Introducing a novel class of nonlinear ensemble filters using measure transport theory.
result Recovery of a constrained Kalman filter for Gaussian settings and combination with regularization techniques.
Convolutional Bayesian filtering generalizes state estimation by incorporating inequality conditions.
problem Standard Bayesian filtering assumes exact conditional probabilities, limiting its applicability.
method Introducing inequality conditions transforms conditional probabilities into convolutional forms, expanding the filtering framework.
result Convolutional Bayesian filtering encompasses standard Bayesian filtering and allows for more nuanced model consideration.
Paper learns to rotate filters for group convolutions.
problem Difficult to rotate 3x3 filters on pixel grids.
method Learn filter basis and rotation-invariant coefficients; switch basis for rotation.
result Produces feature maps insensitive to input rotations.
EnSF improves accuracy in tracking high-dimensional nonlinear systems.
problem Low accuracy in high-dimensional, nonlinear filtering problems.
method Score-based diffusion model, mini-batch Monte Carlo estimator.
result EnSF outperforms state-of-the-art methods in tracking high-dimensional systems.
Non-linear filter aggregation improves image denoising.
problem Efficiently denoise images with complex noise patterns.
method Non-linear aggregation of preliminary filters using a new proximity metric.
result The aggregated filters outperform individual filters in denoising.
Improved Kalman filter for Stiefel manifold measurements.
problem Improving accuracy in measurements on Stiefel manifolds.
method Generalization of extended Kalman filter for Stiefel manifold-valued measurements.
result Significant improvement over raw measurements.
Latent FxLMS accelerates ANC by adapting along low-dimensional filter weights.
problem Improving active noise control with neural adaptive filters.
method Training an auto-encoder on filter coefficients, constraining weights to latent variables, and updating in latent space.
result Latent FxLMS converges in fewer steps with comparable error to standard FxLMS.
In this paper we introduce a projection method for the space of probability distributions based on the differential geometric approach to statistics. This method is based on a direct L2 metric as opposed to the usual Hellinger distance and the related Fisher Information metric. We explain how this apparatus can be used…