Paper accelerates nonlinear mapping in online systems with lower time complexity.
problem Speeding up nonlinear mapping in online systems.
method Integrates an acceleration module into Dendrite Net (DD) to reduce time complexity.
result DD with AC has lower time complexity while maintaining nonlinear mapping and system identification properties.
SUOD accelerates unsupervised outlier detection for large datasets.
problem Scalability issues in unsupervised outlier detection for high-dimensional datasets.
method Three-module acceleration framework: Random Projection, Balanced Parallel Scheduling, Pseudo-supervised Approximation.
result SUOD significantly improves efficiency and scalability in outlier detection.
Many prediction problems, such as those that arise in the context of robotics, have a simplifying underlying structure that, if known, could accelerate learning. In this paper, we present a strategy for learning a set of neural network modules that can be combined in different ways. We train different modular structure…
New CNN architecture improves pediatric image segmentation by homogenizing pose and size.
problem Challenges in segmenting pediatric images due to pose and size heterogeneity.
method Spatial Transformer Network (STN) for pose and scale invariance, combined with UNet for segmentation.
result Improved pediatric segmentation, especially renal tumor delineation, with accelerated processing.
Spectrum selectively trains LLMs based on SNR to save resources.
problem Efficiently training large language models with limited resources.
method Targeting layer modules based on SNR for selective training.
result Spectrum achieves similar performance to full fine-tuning but with reduced VRAM usage.
Skeinformer accelerates self-attention for long sequences with linear complexity.
problem Efficiency of Transformer models in processing long sequences.
method Matrix sketching and column sampling to reduce quadratic complexity to linear.
result Skeinformer outperforms alternatives with smaller time/space footprint.
aweSOM accelerates SOM clustering for large datasets.
problem Scalability issues in existing SOM implementations for large, multidimensional data.
method CPU/GPU-accelerated Self-organizing Maps (SOM) with ensemble stacking.
result 10-100x speed up and improved memory efficiency for large datasets.
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.
Obtaining accurate and reliable images from low-dose computed tomography (CT) is challenging. Regression convolutional neural network (CNN) models that are learned from training data are increasingly gaining attention in low-dose CT reconstruction. This paper modifies the architecture of an iterative regression CNN, BC…
The Hawkes process (HP) has been widely applied to modeling self-exciting events including neuron spikes, earthquakes and tweets. To avoid designing parametric triggering kernel and to be able to quantify the prediction confidence, the non-parametric Bayesian HP has been proposed. However, the inference of such models …
This paper proposes an alternative to E2E training for deep networks, reducing memory footprint.
problem High GPUs memory footprint in end-to-end training of deep networks.
method Locally supervised learning with information propagation loss to avoid information collapse.
result The proposed method achieves competitive performance with less than 40% memory footprint compared to E2E training.
While accelerators such as GPUs have limited memory, deep neural networks are becoming larger and will not fit with the memory limitation of accelerators for training. We propose an approach to tackle this problem by rewriting the computational graph of a neural network, in which swap-out and swap-in operations are ins…
Pykg2vec is an open-source Python library for learning the representations of the entities and relations in knowledge graphs. Pykg2vec's flexible and modular software architecture currently implements 16 state-of-the-art knowledge graph embedding algorithms, and is designed to easily incorporate new algorithms. The goa…
Enhances deep neural networks for MRI reconstruction by increasing expressivity.
problem Balancing network complexity and performance in deep learning MRI reconstruction.
method Geometric approach using bootstrapping and subnetwork aggregation with attention module.
result Significant improvement in MRI reconstruction performance with minimal complexity increase.
New Convolutional Unit improves Batch Whitening performance.
problem Improving the efficiency and effectiveness of Batch Whitening.
method Proposes a new Convolutional Unit that aligns with Batch Whitening theory and empirically analyzes the original Convolutional Unit.
result Significantly improved performance on multiple image classification datasets.
This paper presents a module of vehicle reidentification based on make/model and color classification. It could be used by the Automated Vehicular Surveillance (AVS) or by the fast analysis of video data. Many of problems, that are related to this topic, had to be addressed. In order to facilitate and accelerate the pr…
In classical Hawkes process, the baseline intensity and triggering kernel are assumed to be a constant and parametric function respectively, which limits the model flexibility. To generalize it, we present a fully Bayesian nonparametric model, namely Gaussian process modulated Hawkes process and propose an EM-variation…
Classifies modules of surface-knots in terms of their properties.
problem Characterizing modules of surface-knots in terms of their properties.
method Using homology and covering spaces, the reduced first module is characterized.
result The reduced first module for every genus g is characterized in terms of properties of a finitely generated module.
A fast method for learning MZI parameters in optical neural networks.
problem Time-consuming learning of MZI parameters in optical neural networks.
method Customized complex-valued derivatives and a chain rule for Wirtinger derivatives, incorporated into a function module.
result 20 times faster learning compared to conventional AD in MNIST task.
Expanding the receptive field to capture large-scale context is key to obtaining good performance in dense prediction tasks, such as human pose estimation. While many state-of-the-art fully-convolutional architectures enlarge the receptive field by reducing resolution using strided convolution or pooling layers, the mo…
Accelerates Riemannian gradient methods with extrapolation.
problem Optimizing functions on manifolds efficiently.
method Extrapolating iterates in Riemannian gradient descent.
result Achieves optimal convergence rate and computational advantage.
Proves finiteness and holonomicity of skein modules for 3-manifolds.
problem Finiteness and holonomicity of skein modules for 3-manifolds.
method Defining skein transfer bimodules and using q-analogues of D-module theory.
result Internal skein modules are holonomic modules over the internal skein algebra of the boundary.
Defines super projective modules and explores their properties.
problem Exploring the geometric-algebraic link in super geometry.
method Defined and explored super projective modules over supersmooth functions.
result Module of vector fields over a supersphere is a super projective module.
We introduce higher skein modules of links generalizing the Conway skein module. We show that these modules are closely connected to the HOMFLY polynomial.
A new method to derive presentations of skein modules is developed. For the case of homotopy skein modules it will be shown how the topology of a 3-manifold is reflected in the structure of the module. The freeness problem for q-homotopy skein modules is solved, and a natural skein module related to linking numbers is …
Neural Module Networks, originally proposed for the task of visual question answering, are a class of neural network architectures that involve human-specified neural modules, each designed for a specific form of reasoning. In current formulations of such networks only the parameters of the neural modules and/or the or…
Paper compares skein modules to Kauffman bracket modules.
problem Comparing skein modules to Kauffman bracket modules.
method Using skein relations and Reshetikhin-Turaev model.
result Resolved the problem of comparing skein modules to Kauffman bracket modules.
Skein modules are the main objects of an algebraic topology based on knots (or position). In the same spirit as Leibniz we would call our approach "algebra situs." When looking at the panorama of skein modules we see, past the rolling hills of homologies and homotopies, distant mountains - the Kauffman bracket skein mo…
Generalized Steinberg module presentation for Gaussian and Eisenstein integers.
problem Presenting Steinberg modules for specific number rings.
method Generalization of Bykovskii's presentation to Gaussian and Eisenstein integers.
result Generalization does not yield a presentation for all Euclidean number rings.
We investigate the combination of actor-critic reinforcement learning algorithms with uniform large-scale experience replay and propose solutions for two challenges: (a) efficient actor-critic learning with experience replay (b) stability of off-policy learning where agents learn from other agents behaviour. We employ …
Accelerates optimization in asynchronous systems with sparse updates.
problem Optimizing finite-sum objectives in asynchronous lock-free environments.
method New accelerated SVRG variant with sparse updates.
result Achieves optimal incremental gradient complexity.
Enhanced Alexander module detects linking numbers in links.
problem Detecting linking numbers in links using Alexander modules.
method Defining and singling out meridians and longitudes in reduced Alexander modules.
result The enhanced Alexander module determines all linking numbers.
We define 2-crossed module bundle 2-gerbes related to general Lie 2-crossed modules and discuss their properties. A 2-crossed module bundle 2-gerbe over a manifold is defined in terms of a so called 2-crossed module bundle gerbe, which is a crossed module bundle gerbe equipped with an extra sructure. It is shown that s…
Combinatorial approach to compute satellite knot invariants using graph theory.
problem Computing knot invariants for satellite knots using bordered Heegaard Floer homology.
method Construct weighted A∞-modules using decorated planar graphs and prove their isomorphism. result Combinatorial proof of A∞ structure relations for the constructed modules. The paper explores how gradient descent trains associative memories, revealing oscillations and convergence issues.
problem Training dynamics of associative memories in overparameterized and underparameterized settings.
method Reduction to particle system dynamics, theory, and experiments.
result Oscillatory transitory regimes and benign loss spikes in overparameterized settings, suboptimal memorization in underparameterized settings.
We analyze Riemannian accelerated methods using a new framework.
problem Understanding Riemannian accelerated gradient methods.
method Riemannian A-HPE framework, focusing on Euclidean A-HPE insights and metric distortion control.
result Characterization of acceleration for various Riemannian methods.
Super-acceleration of gradient descent with momentum improves loss function minimization.
problem Minimizing loss functions in machine learning.
method Extending Nesterov acceleration by using gradients at multiple steps ahead.
result Super-acceleration of the momentum algorithm is beneficial for various loss landscapes and tasks.
New sl(2) action defined on a mathematical module.
problem No specific problem stated; focuses on mathematical construction.
method Construction of sl(2)-action on equivariant skein lasagna module.
result Infinitesimal sl(2)-symmetries constructed.
PF-LaCG removes the need for knowing smoothness and strong convexity parameters for locally accelerated CG.
problem Locally accelerated CG requires knowledge of smoothness and strong convexity parameters.
method Parameter-Free Locally Accelerated CG (PF-LaCG) algorithm.
result PF-LaCG achieves local acceleration without requiring knowledge of smoothness and strong convexity parameters.
Introduces admissible skein modules for non-semisimple categories.
problem No specific problem stated; generalization of Kauffman skein algebra.
method Introduces admissible skein modules associated to ideals in pivotal categories.
result These modules generalize Kauffman skein algebra and relate to quantum invariants.
Enhances knot and link invariants using quandle modules.
problem Distinguishing knots and links using polynomial invariants.
method Integrates quandle modules into the quandle coloring quiver.
result The enhanced invariant distinguishes knots and links.
Studies modules over a category of Jacobi diagrams in handlebodies.
problem Understanding modules over a specific category of Jacobi diagrams.
method Generalizes adjunctions and studies subquotient modules.
result Generalizes adjunctions between modules and Casimir Lie algebra modules.
This paper generalizes L2 cohomology theory for complex manifolds.
problem Developing a L2 cohomology theory for Hodge modules on infinite covering spaces.
method Formulating a conjectural generalization of L2-Mixed Hodge structures using Saito's Mixed Hodge Modules.
result Partial results in the conjectural generalization of L2-Mixed Hodge structures.
Let {T1,…,Tn} be a set of n commuting bounded linear operators on a Hilbert space H. Then the n-tuple (T1,…,Tn) turns H into a module over C[z1,…,zn] in the following sense: \[\mathbb{C}[z_1, \ldots, z_n] \times \mathcal{H} \raro \clh, \quad \quad …
Study Kauffman bracket skein modules of Seifert fibered spaces.
problem Understanding the structure of Kauffman bracket skein modules.
method Investigate spanning sets and module structure.
result Kauffman bracket skein modules are finitely generated.
Accelerates coordinate descent methods for machine learning problems.
problem Slowness of coordinate descent methods in machine learning.
method Extrapolation-based accelerated coordinate descent.
result Significant speed-up in practice compared to existing methods.
Formula for interleaving distance of rectangle persistence modules.
problem Calculating distances between rectangle persistence modules.
method Formulas based on rectangle geometry, extended to decomposable modules.
result Closed formulas for interleaving and bottleneck distances.
Introduces Floer lasagna modules using link Floer homology.
problem No specific problem stated; focuses on new mathematical concept.
method Inspired by skein lasagna module, uses link Floer homology.
result Computes Floer lasagna modules for specific 4-manifolds.