arXiv research

Reduced modeling of a computationally demanding dynamical system aims at approximating its trajectories, while optimizing the trade-off between accuracy and computational complexity. In this work, we propose to achieve such an approximation by first embedding the trajectories in a reproducing kernel Hilbert space (RKHS…

Existing methods for reducing the computational burden of neural networks at run-time, such as parameter pruning or dynamic computational path selection, focus solely on improving computational efficiency during inference. On the other hand, in this work, we propose a novel method which reduces the memory footprint and…

This paper optimizes AI inference on edge devices with reduced communication and computation costs.

We study the maps induced on link Floer homology by elementary decorated link cobordisms. We compute these for births, deaths, stabilizations, and destabilizations, and show that saddle cobordisms can be computed in terms of maps in a decorated skein exact triangle that extends the oriented skein exact triangle in knot…

Secure neural network inference on untrusted platforms using holographic reduced representations.

A new method reduces deep learning training costs by 92%.

Braid monodromy is an important tool for computing invariants of curves and surfaces. In this paper, the \emph{rectangular braid diagram (RBD)} method is proposed to compute the braid monodromy of a completely reducible $n$-gonal curve, i.e. the curves in the form $(y-y_1(x))...(y-y_n(x))=0$ where $n\in \mathbb{Z}^{+}$…

This work reduces computation cost for on-device CNN training.

Improves distributed SGD convergence speed with reduced computation load.

Reduced reservoir size for faster edge computing.

A new Bayesian image segmentation algorithm is proposed by combining a loopy belief propagation with an inverse real space renormalization group transformation to reduce the computational time. In results of our experiment, we observe that the proposed method can reduce the computational time to less than one-tenth of …

Paper proposes efficient BNN inference flow to reduce computation and memory costs.

The paper extends a variance gamma model to quadratic functions, reducing arbitrage and computational costs.

Recurrent neural networks can be difficult to train on long sequence data due to the well-known vanishing gradient problem. Some architectures incorporate methods to reduce RNN state updates, therefore allowing the network to preserve memory over long temporal intervals. To address these problems of convergence, this p…

A computational theory reduces agent evaluation errors and speeds up processes.

In this paper we compute the reduced HOMFLY-PT homologies of the Conway and the Kinoshita-Terasaka knots and show that they are isomorphic.

Computer experiments reveal complex knots that don't simplify.

A faster Wasserstein k-means algorithm for histogram data reduces computation and maintains clustering quality.

Efficiently discovers Bayesian network structure with reduced memory usage.

Unreduced PDs can perform similarly to reduced PDs in machine learning tasks.

LightOn OPUs accelerate randomized numerical linear algebra, reducing computational costs.

Optical ESNs enable flexible, efficient machine learning with reduced energy.

Study projective derivative cocycles for circle diffeomorphisms.

Cost-aware SBI reduces expensive simulations in complex models.

Sparsity helps reduce diffusion model costs.

Neural networks have proven to be extremely powerful tools for modern artificial intelligence applications, but computational and storage complexity remain limiting factors. This paper presents two compatible contributions towards reducing the time, energy, computational, and storage complexities associated with multil…

Chebyshev technique reduces FRTB-IMA equity autocallables computation costs by 90%.

New methods accelerate NCGP inference by trading computation for uncertainty.

Ball k-means reduces point-centroid distance computations for faster k-means clustering.

We proved by computer enumeration that the Jones polynomial distinguishes the unknot for knots up to 22 crossings. Following an approach of Yamada, we generated knot diagrams by inserting algebraic tangles into Conway polyhedra, computed their Jones polynomials by a divide-and-conquer method, and tested those with triv…

One-step learning in crosspoint memory reduces computation time.

The calculation of minimum energy paths for transitions such as atomic and/or spin re-arrangements is an important task in many contexts and can often be used to determine the mechanism and rate of transitions. An important challenge is to reduce the computational effort in such calculations, especially when ab initio …

We propose the use of incomplete dot products (IDP) to dynamically adjust the number of input channels used in each layer of a convolutional neural network during feedforward inference. IDP adds monotonically non-increasing coefficients, referred to as a "profile", to the channels during training. The profile orders th…

Quaternion self-attention reduces computational cost and improves performance.

Popular deep neural networks (DNNs) spend the majority of their execution time computing convolutions. The Winograd family of algorithms can greatly reduce the number of arithmetic operations required and is present in many DNN software frameworks. However, the performance gain is at the expense of a reduction in float…

A new method reduces communication costs in distributed learning.

This paper develops coding techniques to reduce the running time of distributed learning tasks. It characterizes the fundamental tradeoff to compute gradients (and more generally vector summations) in terms of three parameters: computation load, straggler tolerance and communication cost. It further gives an explicit c…

We compute the categorified sl(N) link invariants as defined by Khovanov and Rozansky, for various links and values of N. This is made tractable by an algorithm for reducing tensor products of matrix factorisations to finite rank, which we implement in the computer algebra package Singular.

Feedforward computation, such as evaluating a neural network or sampling from an autoregressive model, is ubiquitous in machine learning. The sequential nature of feedforward computation, however, requires a strict order of execution and cannot be easily accelerated with parallel computing. To enable parallelization, w…

A new method reduces computational cost for gene expression inference in large microarray data sets.

New method models PDEs from noisy, limited data.

Enhances LMC for log-concave sampling, reducing computational cost.

In this brief, we improve the Broad Learning System (BLS) [7] by reducing the computational complexity of the incremental learning for added inputs. We utilize the inverse of a sum of matrices in [8] to improve a step in the pseudoinverse of a row-partitioned matrix. Accordingly we propose two fast algorithms for the c…

A new method reduces feature screening cost from $O(np)$ to $O(\sqrt{n}p)$.

The scalability of submodular optimization methods is critical for their usability in practice. In this paper, we study the reducibility of submodular functions, a property that enables us to reduce the solution space of submodular optimization problems without performance loss. We introduce the concept of reducibility…

Paper reduces neural network complexity for image classification.

Dr. of Crosswise proposes a new architecture to reduce over-parametrization in Neural Networks. It introduces an operand for rapid computation in the framework of Deep Learning that leverages learned weights. The formalism is described in detail providing both an accurate elucidation of the mechanics and the theoretica…

Study Anosov representations of reducible suspensions of hyperbolic groups.