Paper introduces a new method for Transformers with linear complexity.
problem No efficient relative positional encoding for linear Transformer models.
method Stochastic Positional Encoding (SPE) that replaces classical RPE.
result SPE behaves like RPE and performs well on benchmarks.
Stochastic encoders outperform deterministic ones in 'perfect perceptual quality'.
problem Understanding when stochastic encoders outperform deterministic ones.
method Provided a toy example to illustrate performance.
result Stochastic encoders can significantly outperform deterministic ones in 'perfect perceptual quality'.
BiPE blends intra-segment and inter-segment encodings for better length extrapolation.
problem Improving length extrapolation in language models.
method Bilevel Positional Encoding (BiPE) that separates intra-segment and inter-segment encodings.
result BiPE enhances length extrapolation across various text modalities.
STRING improves 2D and 3D position encodings for better performance.
problem Efficient and accurate position encoding for 2D and 3D applications.
method STRING extends Rotary Position Encodings with a unifying theoretical framework, maintaining translation invariance and low computational cost.
result STRING shows substantial gains in open-vocabulary object detection and robotics.
Randomized positional encodings boost transformer performance on longer sequences.
problem Transformers struggle with generalizing to sequences of arbitrary length.
method Introduced randomized positional encodings that simulate longer sequences and randomly select positions.
result Randomized positional encodings increase test accuracy by 12.0% on average for sequences of unseen length.
New method for Transformer models to encode position information without sequential bias.
problem Lack of flexible and learnable position encoding for Transformer models.
method Continuous dynamical model to learn position encoding.
result Consistent improvements over baselines in various NLP tasks.
Signed graphs encode positive (attractive) and negative (repulsive) relations between nodes. We extend spectral clustering to signed graphs via the one-parameter family of Signed Power Mean Laplacians, defined as the matrix power mean of normalized standard and signless Laplacians of positive and negative edges. We pro…
A novel capsule network model improves surrogate modeling and uncertainty quantification from sparse data.
problem Surrogate modeling and uncertainty quantification of systems from sparse data.
method Adapted Capsule Network (CapsNet) architecture into image-to-image regression encoder-decoder network.
result The proposed approach accurately, efficiently, and robustly predicts responses for arbitrary diffusion fields.
Efficiently accelerates attention calculation for Transformers with relative positional encoding.
problem Quadratic complexity of attention in long sequences.
method Kernelized attention with Fast Fourier Transform (FFT) for RPE.
result Achieves O(n log n) time complexity, mitigates training instability, and outperforms other models.
Transformer struggles with arithmetic length but improves with explicit structure encoding.
problem Transformers fail to generalize length in arithmetic tasks.
method Explicitly encoding structural symmetries via modified number formatting and custom positional encodings.
result Transformer can generalize up to 50-digit numbers without additional data.
Paper proposes LCP for structural encodings, outperforming existing methods.
problem Improving Graph Neural Networks performance through effective structural encodings.
method Geometric perspective, Local Curvature Profiles (LCP) for structural encodings, combining with global positional encodings, comparing with rewiring techniques.
result LCP significantly outperforms existing structural encodings and combining LCP with global positional encodings improves performance.
Deep learning models have shown state-of-the-art performance in many inverse reconstruction problems. However, it is not well understood what properties of the latent representation may improve the generalization ability of the network. Furthermore, limited models have been presented for inverse reconstructions over ti…
New analysis shows RPE-based Transformers can't approximate all functions.
problem Understanding the limitations of RPE-based Transformers in approximating continuous functions.
method Mathematical analysis and development of a novel attention module (URPE) to overcome limitations.
result RPE-based Transformers can't approximate all continuous sequence-to-sequence functions, even with depth and width.
GTA improves transformer-based NVS models by encoding geometric structure.
problem Suboptimal positional encoding for 3D vision tasks.
method Geometry-aware attention mechanism encoding geometric structure of tokens.
result GTA improves learning efficiency and performance of NVS models.
Unified framework analyzes and compares RFF and RoPE PEs for music generation.
problem Efficiently modeling music generation with positional encodings.
method Kernel methods to analyze and compare RFF and RoPE PEs.
result RoPEPool outperforms other methods in melody harmonization.
Binary encoding enables neural networks to extrapolate periodic functions.
problem Extrapolating periodic functions without prior knowledge of their form.
method Normalized Base-2 Encoding (NB2E) for continuous numerical values.
result MLPs using NB2E can successfully extrapolate diverse periodic signals.
The variational autoencoder (VAE) imposes a probabilistic distribution (typically Gaussian) on the latent space and penalizes the Kullback--Leibler (KL) divergence between the posterior and prior. In NLP, VAEs are extremely difficult to train due to the problem of KL collapsing to zero. One has to implement various heu…
The paper ensures positivity of solutions to stochastic equations with positive initial data.
problem Ensuring positivity of solutions to stochastic equations with positive initial data.
method Providing sufficient conditions on coefficients for positivity of mild solutions.
result Sufficient conditions for positivity of solutions to stochastic equations.
Graph attention networks improve performance on heterogeneous graphs.
problem Complex performance of GNNs on heterogeneous graphs.
method Integrating positional encodings into graph attention networks.
result Graph attention networks excel in node classification and link prediction.
We simplify Volterra process predictions by reducing dimensionality and using a tailored deep learning model.
problem Predicting the conditional law of Volterra processes with stochastic volatility is challenging due to high dimensionality and non-smoothness.
method We developed a stable dimension reduction technique onto a low-dimensional statistical manifold of non-positive curvature and introduced a sequentially deep learning model tailored to this geometry.
result Our model can approximate the conditional law of Volterra processes with approximation rates achievable only with very large networks.
SFM resolves small-scale physics challenges in weather data.
problem Challenges in super-resolving small-scale details in physical sciences like weather.
method Encoding inputs to a latent base distribution, flow matching for stochastic details, adaptive noise scaling.
result SFM framework significantly outperforms existing methods.
Stochastic encoding improves gender classification of brain networks from UK Biobank data.
problem Complexity and bias in interpreting deep learning models of brain connectivity.
method Stochastic encoding in ensemble of CNNs, multivariate balancing algorithm.
result AUROC of 0.8459, with resting-state data more accurate than task data.
This paper explores GNN functions on random graphs, highlighting the importance of node Positional Encodings.
problem Understanding the expressive power of GNNs on large random graphs.
method General convergence notions, input node features, and Positional Encodings (PEs).
result GNNs can converge to certain functions on large random graphs, emphasizing the role of PEs.
Proposes flexible auto-encoders for varying data dimensions.
problem Fixed latent dimensions limit data flexibility.
method Stochastic bottleneck with weighted dropouts.
result Seamless variable dimensionality reduction with high performance.
New analysis shows PE in Transformers increases generalization gap and vulnerability.
problem Understanding the impact of PE on Transformer generalization and robustness.
method Generalization analysis and adversarial Rademacher bounds for a single-layer Transformer with trainable PE.
result PE systematically enlarges the generalization gap and makes models more vulnerable to attacks.
PE-SVI reduces SVI inference complexity by finding a suitable start point.
problem Complex posterior inference in graphical models leads to suboptimal learning.
method PE-SVI uses a pseudo-encoded start point to reduce gradient steps and step sizes.
result PE-SVI achieves the same ELBo objective as SVI with less than 1% of the required steps.
ST-MAML tackles task ambiguity in meta-learning by encoding tasks with stochastic representations.
problem Handling tasks from multiple distributions is challenging for meta-learning due to task ambiguity.
method ST-MAML uses a stochastic neural network module to encode tasks and propagate task representations to revise input variable encoding.
result ST-MAML matches or outperforms state-of-the-art methods on various tasks.
PE-GQNN improves spatial data prediction and uncertainty quantification.
problem Poor calibration of predictive distributions in spatial data models.
method Combines PE-GNNs with Quantile Neural Networks and recalibration techniques.
result PE-GQNN outperforms existing methods in predictive accuracy and uncertainty quantification.
New method for directed graphs using learnable spectral positional encodings.
problem Challenges in magnetic Laplacians and unitary gauge invariance for directed graphs.
method Learnable spectral PEs of the form hθ(Aq)R, computed in Hermitian block Krylov subspace.
result Gauge-invariant and computationally efficient solution for directed graphs.
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.
A bandit algorithm reduces regret in noisy, communication-constrained feedback.
problem Distributed stochastic multi-armed bandit with noisy, communication-constrained feedback.
method Proposes a multi-phase bandit algorithm, UE-UCB++, that matches an information-theoretic lower bound.
result Matches an information-theoretic lower bound of Ω(√(KT/σ²)) on the minimax regret.
A new method for spectral positional encodings in directed graphs using Hermitian block Krylov subspaces.
problem Challenges in spectral positional encodings for directed graphs, including computational complexity and gauge invariance issues.
method Learnable spectral positional encodings of the form hθ(Aq)R, computed in a Hermitian block Krylov subspace from sparse matrix-vector products. result The method is gauge-invariant and converges to the exact eigendecomposition oracle as the depth grows.
Improved graph embedding through refined linear transformation and community recovery.
problem Identifying meaningful latent communities in graph data.
method Refined graph encoder embedding via linear transformation, self-training, and latent community recovery.
result Improved vertex embedding and better decision boundaries for vertex classification.
Paper formalizes and analyzes a new bound for variational inference.
problem Lack of theoretical guarantees in variational algorithms.
method Introduces VR-IWAE bound, a generalization of IWAE.
result VR-IWAE bound leads to unbiased gradient estimators.
New techniques improve channel prediction in noisy wireless systems.
problem Predicting channels in wireless communication systems from noisy observations.
method Adapted sequence-to-sequence models and transformers with reverse positional encoding and reversed encoder outputs.
result Improved robustness and relationship capture in channel prediction models.
VCAE improves autoencoder quality on MNIST and CelebA.
problem Overfitting and poor generative/reconstruction quality in autoencoders.
method Proposes variance-constrained autoencoder (VCAE) to enforce variance constraint on latent distribution.
result VCAE outperforms Wasserstein Autoencoder and Variational Autoencoder in quality.
Method learns model for unknown stochastic system from data.
problem Modeling unknown stochastic dynamical systems.
method Autoencoder approach using deep neural networks (DNNs).
result Decoder serves as a predictive model for unknown stochastic systems.
Paper shows L∞-positivity and stochastic completeness are equivalent.
problem Analyzing L∞-positivity preserving property and stochastic completeness. method Using monotone approximation results for distributional solutions of −Δ+1≥0. result The L∞-positivity preserving property is equivalent to stochastic completeness. Conformer encoder reverses sequence in time dimension, affecting decoder training.
problem Reversal of sequence in Conformer encoder impacts decoder training.
method Analyzed initial behavior of decoder cross-attention and proposed methods to avoid flipping.
result Self-attention module of Conformer starts dominating, allowing only reversed information to pass.
Brain networks in fMRI are typically identified using spatial independent component analysis (ICA), yet mathematical constraints such as sparse coding and positivity both provide alternate biologically-plausible frameworks for generating brain networks. Non-negative Matrix Factorization (NMF) would suppress negative BO…
Affine spiking neural networks learn efficiently and generalize well.
problem Learning with spiking neural networks, especially with positive weights.
method Affine encoders and decoders, continuous parameter dependence, gradient-based training.
result Affine spiking neural networks can approximate shallow ReLU networks and generalize well.
Maximum principle proves positivity of forward rates in stochastic models.
problem Proving positivity of forward rates in stochastic models.
method Maximum principle for mild solutions to SPDEs with Lipschitz coefficients and Wiener noise.
result Sufficient conditions for positivity of forward rates in the Heath-Jarrow-Morton model.
Study SGD dynamics in sequence models, revealing training phases and influence of sequence length.
problem Understanding SGD in sequence models like attention networks.
method Derived closed-form population loss and analyzed SGD dynamics for SSI models.
result Two distinct training phases: escape from uninformative initialization and alignment with target subspace.
Financial fraud detection in digital banking requires reasoning over multiple heterogeneous event streams.
problem Financial fraud detection in digital banking requires reasoning over multiple heterogeneous event streams.
method Multi-Stream Fraud Transformer (MSFT) architecture that encodes each event stream with independent Transformer encoders and fuses their representations through configurable mechanisms.
result Sequence models significantly outperform gradient-boosted trees operating on aggregated features.
Deterministic training improves generative autoencoder performance.
problem Stochastic training limits generative autoencoder performance.
method Invertible layers for deterministic training.
result AEFs outperform VAEs in log-likelihood and sample quality.
This paper proposes grid cells encode position via a conformal isometric embedding of 2D physical space.
problem Hexagonal grid firing patterns in grid cells.
method Learning a distance-preserving position embedding in neural space using a recurrent neural network.
result The conformal isometric embedding of 2D physical space into neural space explains hexagonal grid firing patterns.
A new model encodes distances and topology in latent variables.
problem Modeling dissimilarity data with latent variables and invariances.
method Isometric Gaussian Process Latent Variable Model using Riemannian geometry and variational inference.
result The model can encode invariances in learned manifolds.
New insights into how encoder-decoder networks generate attention matrices.
problem Understanding how encoder-decoder networks use attention matrices.
method Decomposing hidden states into temporal and input-driven components.
result Attention matrices are formed based on task requirements, not architecture type.