KKnowledgeTrendResearch Network
Research institution

Google (United States)

US

12 researchers0 verified5 linked papers42,885 indexed citations

Researchers

Public research profiles associated with Google (United States).

Research from this institution

Publications linked through researcher authorship records.

2020 · Nature · 23,187 citations

Array programming with NumPy

Abstract Array programming provides a powerful, compact and expressive syntax for accessing, manipulating and operating on data in vectors, matrices and higher-dimensional arrays. NumPy is the primary array programming library for the Python language. It has an essential role in research analysis pipelines in fields as diverse as physics, chemistry, astronomy, geoscience, biology, psychology, materials science, engineering, finance and economics. For example, in astronomy, NumPy was an important part of the software stack used in the discovery of gravitational waves 1 and in the first imaging of a black hole 2 . Here we review how a few fundamental array concepts lead to a simple and powerful programming paradigm for organizing, exploring and analysing scientific data. NumPy is the foundation upon which the scientific Python ecosystem is constructed. It is so pervasive that several projects, targeting audiences with specialized needs, have developed their own NumPy-like interfaces and array objects. Owing to its central position in the ecosystem, NumPy increasingly acts as an interoperability layer between such array computation libraries and, together with its application programming interface (API), provides a flexible framework to support the next decade of scientific and industrial analysis.

2012 · IEEE Signal Processing Magazine · 10,380 citations

Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research Groups

Most current speech recognition systems use hidden Markov models (HMMs) to deal with the temporal variability of speech and Gaussian mixture models (GMMs) to determine how well each state of each HMM fits a frame or a short window of frames of coefficients that represents the acoustic input. An alternative way to evaluate the fit is to use a feed-forward neural network that takes several frames of coefficients as input and produces posterior probabilities over HMM states as output. Deep neural networks (DNNs) that have many hidden layers and are trained using new methods have been shown to outperform GMMs on a variety of speech recognition benchmarks, sometimes by a large margin. This article provides an overview of this progress and represents the shared views of four research groups that have had recent successes in using DNNs for acoustic modeling in speech recognition.

2019 · arXiv (Cornell University) · 5,012 citations

EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks

Convolutional Neural Networks (ConvNets) are commonly developed at a fixed resource budget, and then scaled up for better accuracy if more resources are available. In this paper, we systematically study model scaling and identify that carefully balancing network depth, width, and resolution can lead to better performance. Based on this observation, we propose a new scaling method that uniformly scales all dimensions of depth/width/resolution using a simple yet highly effective compound coefficient. We demonstrate the effectiveness of this method on scaling up MobileNets and ResNet. To go even further, we use neural architecture search to design a new baseline network and scale it up to obtain a family of models, called EfficientNets, which achieve much better accuracy and efficiency than previous ConvNets. In particular, our EfficientNet-B7 achieves state-of-the-art 84.3% top-1 accuracy on ImageNet, while being 8.4x smaller and 6.1x faster on inference than the best existing ConvNet. Our EfficientNets also transfer well and achieve state-of-the-art accuracy on CIFAR-100 (91.7%), Flowers (98.8%), and 3 other transfer learning datasets, with an order of magnitude fewer parameters. Source code is at https://github.com/tensorflow/tpu/tree/master/models/official/efficientnet.

2022 · arXiv (Cornell University) · 4,301 citations

BNAI, NO-TOKEN, and MIND-UNITY: Pillars of a Systemic Revolution in Artificial Intelligence

AbstractThere is a failure mode in large language models that we do not have a good name for, and thatwe therefore tend not to treat seriously enough. It is not hallucination — the model is not assertingsomething false. It is not refusal — the model answers at length. It is the production of responses thatcarry the complete outward form of careful reasoning while the cognitive work that reasoning issupposed to represent has not, in any meaningful sense, occurred. We call this theatrical compliance,and we argue that it is, in practical terms, more dangerous than either of the failure modes thatcurrently dominate alignment research. This paper identifies the phenomenon, characterizes its fiveprincipal forms, explains the asymmetry that makes it particularly costly in high-stakes settings, andoutlines the design requirements for systems intended to resist it. We do not describe such a systemin detail here. Our goal is to establish theatrical compliance as a research problem in its own rightand to argue that addressing it requires instruments operating at a fundamentally different level ofabstraction than task-level prompting frameworks.Keywords: theatrical compliance, large language models, AI reasoning quality, cognitiveprocess evaluation, prompt engineering, metacognitive systems.

2019 · EPJ Web of Conferences · 5 citations

The Data Ocean Project

Transparent use of commercial cloud resources for scientific experiments is a hard problem. In this article, we describe the first steps of the Data Ocean R&D collaboration between the high-energy physics experiment ATLAS together with Google Cloud Platform, to allow seamless use of Google Compute Engine and Google Cloud Storage for physics analysis. We start by describing the three preliminary use cases that were identified at the beginning of the project. The following sections then detail the work done in the data management system Rucio and the workflow management systems PanDA and Harvester to interface Google Cloud Platform with the ATLAS distributed computing environment, and show the results of the integration tests. Afterwards, we describe the setup and results from a full ATLAS user analysis that was executed natively on Google Cloud Platform, and give estimates on projected costs. We close with a summary and and outlook on future work.