weak supervised learning on ray
play

Weak Supervised Learning on Ray using Snorkel British Sentiment - PowerPoint PPT Presentation

Aug 05, 2023 •133 likes •219 views

Weak Supervised Learning on Ray using Snorkel British Sentiment Analysis with LSTM Using Noisy Crowd Sami (sa894) - R244 Large-Scale Data Processing and Optimisation Snorkel generating training label Generating training labels with weak-

  1. Weak Supervised Learning on Ray using Snorkel British Sentiment Analysis with LSTM Using Noisy Crowd Sami (sa894) - R244 Large-Scale Data Processing and Optimisation

  2. Snorkel – generating training label • Generating training labels with weak- supervision • Apply to unlabelled large quintinites of data • Quantity over quality

  3. Snorkel using Spark (Snark) [1] • Quantity = scales. Scale = Large heavy systems. But how heavy? • Load Data into SparkSQL and Dataframes • Pre-process the data and convert them to Snorkel format • Load batches of the snorkel data into Spark clusters • Distribute the workload using Spark actors to apply labelling functions • Spark is based on Scala.

  4. Ray – Making Snorkel spark • Distributed execution backend optimised for ML tasks • Lightweight actor model, ideal for iterative process • E.g. going through batches of data and apply labelling function. • Extend the Snorkel codebase to allow an easy to use interface that uses Ray

  5. The desired outcome • Comparison between Ray and Spark as a backend in applying labelling functions. • Labelling functions are short lived, Ray should in theory be better, but will it be? • Open source the interface and making it available to public.

  6. Demo - How’s the weather

  7. Sentiments Analysis - The British Version • The most common asked question in the UK. • Tweets sentiment analysis in regards to the weather: Positive, negative, neutral, not related, and unsure. • Snorkel to resolve conflicts in a noisy crowdsourced dataset, from Crowdflower. • Then use the denoised labels for an LSTM sentiments analysis • The idea is to dogfood the interface • Based on tutorial for learning Snorkel [0]

  8. Progress - Somewhere ✔︐ Read the papers and related work • ✔︐ Go through the tutorial for Ray and Snorkel • ✔︐ Figure out the best level of abstraction (inspired by Snark) • ✔︐ write those slides • • Next • Implement RayAnnotationDistribution • Run the tests provided in the Snorkel package • Evaluate the performance of the test • Evaluate the performance of a real application – Sentiment analysis • Write down the results

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

Few-shot learning of weak supervision sources in Snorkel (or, learning weakly supervised weak

Few-shot learning of weak supervision sources in Snorkel (or, learning weakly supervised weak

Few-shot learning of weak supervision sources in Snorkel (or, learning weakly supervised weak supervisors) Jesse Mu Project Outline Replicate Snorkel causal relation extraction system Learn weak supervision sources from tiny sets of

364 views • 12 slides
Learning frameworks Self-supervised learning: (Auto)encoder networks Supervised learning Network

Learning frameworks Self-supervised learning: (Auto)encoder networks Supervised learning Network

Learning frameworks Self-supervised learning: (Auto)encoder networks Supervised learning Network must copy inputs to outputs through Assumes environment specifies correct output (targets) for each input a bottleneck (fewer hidden units)

407 views • 5 slides
A Supertag-Context Model for Weakly-Supervised CCG Parser Learning Dan Garrette U. Washington

A Supertag-Context Model for Weakly-Supervised CCG Parser Learning Dan Garrette U. Washington

A Supertag-Context Model for Weakly-Supervised CCG Parser Learning Dan Garrette U. Washington Chris Dyer CMU Jason Baldridge UT-Austin Noah A. Smith CMU Contributions 1. A new generative model for learning CCG parsers from weak

912 views • 65 slides
Self-Supervised Feature Learning by Learning to Spot Artifacts Wonbin Kim Self-Supervised

Self-Supervised Feature Learning by Learning to Spot Artifacts Wonbin Kim Self-Supervised

Self-Supervised Feature Learning by Learning to Spot Artifacts Wonbin Kim Self-Supervised Learning To exploit different labelings that are freely available besides or within visual data, and to use them as intrinsic reward signals to

387 views • 24 slides
Stacking for supervised learning Stacking for supervised learning Niall Rooney, NIKEL,

Stacking for supervised learning Stacking for supervised learning Niall Rooney, NIKEL,

Stacking for supervised learning Stacking for supervised learning Niall Rooney, NIKEL, University of Ulster 1 Ensemble learning Ensemble learning l Postulate multiple hypotheses to explain the data l Shortcomings of single model learning

681 views • 19 slides
Machine Learning for NLP Supervised Learning Aurlie Herbelot 2019 Centre for Mind/Brain

Machine Learning for NLP Supervised Learning Aurlie Herbelot 2019 Centre for Mind/Brain

Machine Learning for NLP Supervised Learning Aurlie Herbelot 2019 Centre for Mind/Brain Sciences University of Trento 1 Supervised learning Supervised: you know the result of the task you want to perform. Supervised learning mostly

832 views • 61 slides
PCA CS 446 Supervised learning So far, weve done supervised learning: Given (( x i , y i )) ,

PCA CS 446 Supervised learning So far, weve done supervised learning: Given (( x i , y i )) ,

PCA CS 446 Supervised learning So far, weve done supervised learning: Given (( x i , y i )) , find f with f ( x i ) y i . k -nn, decision trees, . . . 1 / 18 Supervised learning So far, weve done supervised learning: Given (( x i , y

1.01k views • 56 slides
Semi-Supervised Learning Maria-Florina Balcan 03/30/2015 Readings: Semi-Supervised Learning.

Semi-Supervised Learning Maria-Florina Balcan 03/30/2015 Readings: Semi-Supervised Learning.

Semi-Supervised Learning Maria-Florina Balcan 03/30/2015 Readings: Semi-Supervised Learning. Encyclopedia of Machine Learning. Jerry Zhu, 2010 Combining Labeled and Unlabeled Data with Co- Training. Avrim Blum, Tom Mitchell. COLT

953 views • 45 slides
RECSM Summer School: Machine Learning for Social Sciences Session 1.3: Supervised Learning and

RECSM Summer School: Machine Learning for Social Sciences Session 1.3: Supervised Learning and

RECSM Summer School: Machine Learning for Social Sciences Session 1.3: Supervised Learning and Model Accuracy Reto West Department of Political Science and International Relations University of Geneva 1 Supervised Learning Supervised

436 views • 42 slides
CSE 258 Lecture 1.5 Web Mining and Recommender Systems Supervised learning Regression

CSE 258 Lecture 1.5 Web Mining and Recommender Systems Supervised learning Regression

CSE 258 Lecture 1.5 Web Mining and Recommender Systems Supervised learning Regression What is supervised learning? Supervised learning is the process of trying to infer from labeled data the underlying function that produced the labels

599 views • 33 slides
CSE 158 Lecture 1.5 Web Mining and Recommender Systems Supervised learning Regression

CSE 158 Lecture 1.5 Web Mining and Recommender Systems Supervised learning Regression

CSE 158 Lecture 1.5 Web Mining and Recommender Systems Supervised learning Regression What is supervised learning? Supervised learning is the process of trying to infer from labeled data the underlying function that produced the labels

751 views • 25 slides
CSE 158 Lecture 1.5 Web Mining and Recommender Systems Supervised learning Regression

CSE 158 Lecture 1.5 Web Mining and Recommender Systems Supervised learning Regression

CSE 158 Lecture 1.5 Web Mining and Recommender Systems Supervised learning Regression What is supervised learning? Supervised learning is the process of trying to infer from labeled data the underlying function that produced the labels

784 views • 33 slides
Machine Learning Lecture 6 Unsupervised Learning with a bit of Supervised Learning Clustering

Machine Learning Lecture 6 Unsupervised Learning with a bit of Supervised Learning Clustering

Machine Learning Lecture 6 Unsupervised Learning with a bit of Supervised Learning Clustering and nearest neighbour classifiers. Justin Pearson 1 2020 1 http://user.it.uu.se/~justin/Teaching/MachineLearning/index.html 1 / 27 Supervised

813 views • 27 slides
Web Mining and Recommender Systems Supervised learning Regression Learning Goals Introduce

Web Mining and Recommender Systems Supervised learning Regression Learning Goals Introduce

Web Mining and Recommender Systems Supervised learning Regression Learning Goals Introduce the concept of Supervised Learning Understand the components (inputs and outputs) of supervised learning problems Introduce linear

1.36k views • 122 slides
THE SUPERVISED LEARNING PROBLEM THE SUPERVISED LEARNING PROBLEM Matthieu R Bloch January 7, 2020

THE SUPERVISED LEARNING PROBLEM THE SUPERVISED LEARNING PROBLEM Matthieu R Bloch January 7, 2020

THE SUPERVISED LEARNING PROBLEM THE SUPERVISED LEARNING PROBLEM Matthieu R Bloch January 7, 2020 1 WHY ML ? WHY ML ? Traditional engineering is top-down We use fundamental principles (mathematics, physics) to build models and abstractions

243 views • 13 slides
Weakly Supervised Classification Weakly Supervised Classification and Robust Learning and Robust

Weakly Supervised Classification Weakly Supervised Classification and Robust Learning and Robust

IIT-H and RIKEN-AIP Joint Workshop on March 15, 2019 Machine Learning and Applications, Hyderabad, India Weakly Supervised Classification Weakly Supervised Classification and Robust Learning and Robust Learning --- Overview of Our Recent

1.06k views • 25 slides
(Enumeration Results for) Signed Graphs Matthias Beck San Francisco State University [John

(Enumeration Results for) Signed Graphs Matthias Beck San Francisco State University [John

(Enumeration Results for) Signed Graphs Matthias Beck San Francisco State University [John Stembridge] math.sfsu.edu/beck Why Graphs? A [directed] graph G = ( V, E ) consists of a node set V V [ V 2 ] an edge set E 2

205 views • 16 slides
USC Stadium PV Team Football Sun-days Team Members: Connor Charney, Dylan Humenik, Evelyn

USC Stadium PV Team Football Sun-days Team Members: Connor Charney, Dylan Humenik, Evelyn

USC Stadium PV Team Football Sun-days Team Members: Connor Charney, Dylan Humenik, Evelyn Bateman, and Howard Holmes Contents Client and Location Technology Review Technological Constraints Socioeconomic Constraints

511 views • 31 slides
Floquet Topological Insulator: UnderstandingFloquet topological insulator in semiconductor

Floquet Topological Insulator: UnderstandingFloquet topological insulator in semiconductor

Introduction Building up our toolbox The Model Analysis Floquet Topological Insulator: UnderstandingFloquet topological insulator in semiconductor quantum wellsby Lindner et al. Condensed Matter Journal Club Caltech February 12 2014

639 views • 45 slides
Internal Audit of Mutual Funds and Portfolio Managers CA Vidya Barje M.P. Chitale & Co.

Internal Audit of Mutual Funds and Portfolio Managers CA Vidya Barje M.P. Chitale & Co.

Internal Audit of Mutual Funds and Portfolio Managers CA Vidya Barje M.P. Chitale & Co. September 17, 2016 Vidya Barje, M.P .Chitale & Co. Internal Audit of Mutual Funds Vidya Barje, M.P Vidya Barje, M.P .Chitale & Co.

1.08k views • 80 slides
Transnational Governance: Trade and Climate links Jaime de Melo FERDI Transnational Governance?

Transnational Governance: Trade and Climate links Jaime de Melo FERDI Transnational Governance?

Transnational Governance: Trade and Climate links Jaime de Melo FERDI Transnational Governance? EUI workshop, Florence, November 1314, 2017 OUTLINE Institutions/Governance, and Trade: slow motion Architecture for a Workable and

490 views • 21 slides
Complex social systems: prospects and problems Nigel Gilbert Centre for Research in Social

Complex social systems: prospects and problems Nigel Gilbert Centre for Research in Social

Complex social systems: prospects and problems Nigel Gilbert Centre for Research in Social Simulation, University of Surrey, UK cress.soc.surrey.ac.uk 1 Wednesday, 14 January 2009 1 Overview Social systems are complex... non-linear

742 views • 36 slides
Welcome to Harrison Assessments Talent Solutions with A Global Integrated Solution 40

Welcome to Harrison Assessments Talent Solutions with A Global Integrated Solution 40

Welcome to Harrison Assessments Talent Solutions with A Global Integrated Solution 40 countries 30 languages 3 million+ users 30+ years of research Harrison Assessments Talent Solutions Canon Rotary International Epson University of

321 views • 28 slides
energy and buildings context Tadj Oreszczyn 2 3 United Kingdom Quotes from ACEEE Energy Score

energy and buildings context Tadj Oreszczyn 2 3 United Kingdom Quotes from ACEEE Energy Score

BSEER approval for your application for a Turing Fellowship UKs non -domestic energy and buildings context Tadj Oreszczyn 2 3 United Kingdom Quotes from ACEEE Energy Score Card 2016 This year the United Kingdom fell behind Germany, Japan,

561 views • 40 slides

More recommend