Search Results

Computing with B-series

6 code implementations23 Nov 2021

We present BSeries. jl, a Julia package for the computation and manipulation of B-series, which are a versatile theoretical tool for understanding and designing discretizations of differential equations.

Numerical Analysis Mathematical Software Numerical Analysis Symbolic Computation

Isomeric trees and the order of Runge--Kutta methods

1 code implementation2 Dec 2021

The conditions for a Runge--Kutta method to be of order $p$ with $p\ge 5$ for a scalar non-autonomous problem are a proper subset of the order conditions for a vector problem.

Numerical Analysis Numerical Analysis 65L05

MediaPipe: A Framework for Building Perception Pipelines

2 code implementations14 Jun 2019

A developer can use MediaPipe to build prototypes by combining existing perception components, to advance them to polished cross-platform applications and measure system performance and resource consumption on target platforms.

Distributed, Parallel, and Cluster Computing

Diverse Beam Search: Decoding Diverse Solutions from Neural Sequence Models

25 code implementations7 Oct 2016

We observe that our method consistently outperforms BS and previously proposed techniques for diverse decoding from neural sequence models.

Image Captioning Machine Translation +4

Fast and Accurate Time Series Classification with WEASEL

1 code implementation26 Jan 2017

On the popular UCR benchmark of 85 TS datasets, WEASEL is more accurate than the best current non-ensemble algorithms at orders-of-magnitude lower classification and training times, and it is almost as accurate as ensemble classifiers, whose computational complexity makes them inapplicable even for mid-size datasets.

Classification General Classification +4

Multiple Instance Learning for Efficient Sequential Data Classification on Resource-constrained Devices

1 code implementation NeurIPS 2018

We propose a method, EMI-RNN, that exploits these observations by using a multiple instance learning formulation along with an early prediction technique to learn a model that achieves better accuracy compared to baseline models, while simultaneously reducing computation by a large fraction.

General Classification Multiple Instance Learning +2

Compositional Attention: Disentangling Search and Retrieval

3 code implementations ICLR 2022

Through our qualitative analysis, we demonstrate that Compositional Attention leads to dynamic specialization based on the type of retrieval needed.

Retrieval

Revised Architecture and Two New Super-Earths in the HD 134606 Planetary System

1 code implementation30 Jan 2024

In addition, we identify a linear trend in the RV time series, and the cause of this acceleration is deemed to be a newly detected sub-stellar companion at large separation.

Earth and Planetary Astrophysics

DPSOM: Deep Probabilistic Clustering with Self-Organizing Maps

2 code implementations3 Oct 2019

We show that DPSOM achieves superior clustering performance compared to current deep clustering methods on MNIST/Fashion-MNIST, while maintaining the favourable visualization properties of SOMs.

Clustering Deep Clustering +4

Chronos: Learning the Language of Time Series

2 code implementations12 Mar 2024

We introduce Chronos, a simple yet effective framework for pretrained probabilistic time series models.

Gaussian Processes Language Modelling +2