Interacting with VW in active learning Nikos Karampatziakis Cloud and Information Sciences Lab Microsoft

NIPS 2013

Active Learning (in VW)

I

Situation: unlabeled (+little labeled) data

Active Learning (in VW)

I

Situation: unlabeled (+little labeled) data

I

Start with whatever data is available

Active Learning (in VW)

I

Situation: unlabeled (+little labeled) data

I

Start with whatever data is available

I

Learner interacts with teacher to learn more

Active Learning (in VW)

I

Situation: unlabeled (+little labeled) data

I

Start with whatever data is available

I

Learner interacts with teacher to learn more In VW interaction consists of

I

Active Learning (in VW)

I

Situation: unlabeled (+little labeled) data

I

Start with whatever data is available

I

Learner interacts with teacher to learn more In VW interaction consists of

I

I

Teacher gives unlabeled example to VW

Active Learning (in VW)

I

Situation: unlabeled (+little labeled) data

I

Start with whatever data is available

I

Learner interacts with teacher to learn more In VW interaction consists of

I

I I

Teacher gives unlabeled example to VW VW decides whether and how much it needs that label

Active Learning (in VW)

I

Situation: unlabeled (+little labeled) data

I

Start with whatever data is available

I

Learner interacts with teacher to learn more In VW interaction consists of

I

I I I

Teacher gives unlabeled example to VW VW decides whether and how much it needs that label Teacher can provide the label (not obliged to)

Active Learning (in VW)

I

Situation: unlabeled (+little labeled) data

I

Start with whatever data is available

I

Learner interacts with teacher to learn more In VW interaction consists of

I

I I I

I

Teacher gives unlabeled example to VW VW decides whether and how much it needs that label Teacher can provide the label (not obliged to)

The result is an importance weighted dataset

Active Learning (in VW)

I

Situation: unlabeled (+little labeled) data

I

Start with whatever data is available

I

Learner interacts with teacher to learn more In VW interaction consists of

I

I I I

Teacher gives unlabeled example to VW VW decides whether and how much it needs that label Teacher can provide the label (not obliged to)

I

The result is an importance weighted dataset

I

No selection bias so can be used any way you like

active interactor.py I

A simple demonstation of how to interact with VW

active interactor.py I I

A simple demonstation of how to interact with VW Assuming

active interactor.py I I

A simple demonstation of how to interact with VW Assuming I

binary classification

active interactor.py I I

A simple demonstation of how to interact with VW Assuming I I

binary classification examples are in human readable form (text)

active interactor.py I I

A simple demonstation of how to interact with VW Assuming I I

I

binary classification examples are in human readable form (text)

Connects to the host:port VW is listening on

active interactor.py I I

A simple demonstation of how to interact with VW Assuming I I

binary classification examples are in human readable form (text)

I

Connects to the host:port VW is listening on

I

Sends any initially available labeled data

active interactor.py I I

A simple demonstation of how to interact with VW Assuming I I

binary classification examples are in human readable form (text)

I

Connects to the host:port VW is listening on

I

Sends any initially available labeled data Sends unlabeled examples one by one

I

active interactor.py I I

A simple demonstation of how to interact with VW Assuming I I

binary classification examples are in human readable form (text)

I

Connects to the host:port VW is listening on

I

Sends any initially available labeled data Sends unlabeled examples one by one

I

I

Waits for VW’s response

active interactor.py I I

A simple demonstation of how to interact with VW Assuming I I

binary classification examples are in human readable form (text)

I

Connects to the host:port VW is listening on

I

Sends any initially available labeled data Sends unlabeled examples one by one

I

I I

Waits for VW’s response If VW does not want the label, sends the next

active interactor.py I I

A simple demonstation of how to interact with VW Assuming I I

binary classification examples are in human readable form (text)

I

Connects to the host:port VW is listening on

I

Sends any initially available labeled data Sends unlabeled examples one by one

I

I I I

Waits for VW’s response If VW does not want the label, sends the next Otherwise, VW’s response includes an importance

active interactor.py I I

A simple demonstation of how to interact with VW Assuming I I

binary classification examples are in human readable form (text)

I

Connects to the host:port VW is listening on

I

Sends any initially available labeled data Sends unlabeled examples one by one

I

I I I I

Waits for VW’s response If VW does not want the label, sends the next Otherwise, VW’s response includes an importance Asks the user for the label

active interactor.py I I

A simple demonstation of how to interact with VW Assuming I I

binary classification examples are in human readable form (text)

I

Connects to the host:port VW is listening on

I

Sends any initially available labeled data Sends unlabeled examples one by one

I

I I I I I

Waits for VW’s response If VW does not want the label, sends the next Otherwise, VW’s response includes an importance Asks the user for the label If user skips, sends the next

active interactor.py I I

A simple demonstation of how to interact with VW Assuming I I

binary classification examples are in human readable form (text)

I

Connects to the host:port VW is listening on

I

Sends any initially available labeled data Sends unlabeled examples one by one

I

I I I I I I

Waits for VW’s response If VW does not want the label, sends the next Otherwise, VW’s response includes an importance Asks the user for the label If user skips, sends the next Otherwise, we have a new labeled weighted example

active interactor.py I I

A simple demonstation of how to interact with VW Assuming I I

binary classification examples are in human readable form (text)

I

Connects to the host:port VW is listening on

I

Sends any initially available labeled data Sends unlabeled examples one by one

I

I I I I I I I

Waits for VW’s response If VW does not want the label, sends the next Otherwise, VW’s response includes an importance Asks the user for the label If user skips, sends the next Otherwise, we have a new labeled weighted example Sends it to VW (causes update).

active interactor.py I I

A simple demonstation of how to interact with VW Assuming I I

binary classification examples are in human readable form (text)

I

Connects to the host:port VW is listening on

I

Sends any initially available labeled data Sends unlabeled examples one by one

I

I I I I I I I I

Waits for VW’s response If VW does not want the label, sends the next Otherwise, VW’s response includes an importance Asks the user for the label If user skips, sends the next Otherwise, we have a new labeled weighted example Sends it to VW (causes update). Saves it to a file, so can quit anytime.

Interacting with VW in active learning - GitHub

Nikos Karampatziakis. Cloud and Information Sciences Lab. Microsoft ... are in human readable form (text). ▷ Connects to the host:port VW is listening on ...

188KB Sizes 17 Downloads 344 Views

Recommend Documents

Cluster-parallel learning with VW - GitHub
´runvw.sh ´ -reducer NONE. Each mapper runs VW. Model stored in /model on HDFS runvw.sh calls VW, used to modify VW ...

Batch optimization in VW via LBFGS - GitHub
Dec 16, 2011 - gt. Problem: Hessian can be too big (matrix of size dxd) .... terminate if either: the specified number of passes over the data is reached.

Cluster-parallel learning with VW - PDFKUL.COM
Goals for future from last year. 1. Finish Scaling up. I want a kilonode program. 2. Native learning reductions. Just like more complicated losses. 3. Other learning algorithms, as interest dictates. 4. Persistent Demonization ...

Interacting with Features in GPlates
See ​www.earthbyte.org/Resources/earthbyte_gplates.html​ for EarthByte data sets. Background. GPlates ... Feature Type box at the top of the window). .... Indonesian Gateway and associated back-arc basins, Earth-Sci. Rev., vol 83, p.

Interacting with Features in GPlates
... the new changes. See the GPlates online manual for further information: ... In this exercise we will learn how to query features. To start with we .... 85-138, 1995. Gaina, C., and R. D. Müller, Cenozoic tectonic and depth/age evolution of the.

Efficient Active Learning with Boosting
compose the set Dn. The whole data set now is denoted by Sn = {DL∪n,DU\n}. We call it semi-supervised data set. Initially S0 = D. After all unlabeled data are labeled, the data set is called genuine data set G,. G = Su = DL∪u. We define the cost

Efficient Active Learning with Boosting
unify semi-supervised learning and active learning boosting. Minimization of ... tant, we derive an efficient active learning algorithm under ... chine learning and data mining fields [14]. ... There lacks more theoretical analysis for these ...... I

Efficient Active Learning with Boosting
[email protected], [email protected]} handle. For each query, a ...... can be easily generalized to batch mode active learn- ing methods. We can ...

Deep Learning with H2O.pdf - GitHub
best-in-class algorithms such as Random Forest, Gradient Boosting and Deep Learning at scale. .... elegant web interface or fully scriptable R API from H2O CRAN package. · grid search for .... takes to cut the learning rate in half (e.g., 10−6 mea

Efficient Active Learning with Boosting
real-world database, which show the efficiency of our algo- rithm and verify our theoretical ... warehouse and internet usage has made large amount of unsorted ...

Vacuum decay in an interacting multiverse
Jun 2, 2016 - to address the cosmological constant problem in the context of string theory [1–3]. ...... where θ = 2Hσt + θ0, with θ0 some constant to fit with the.

"Support vector machine active learning with ... -
"Support vector machine active learning with applications ... The Journal of Machine Learning Research 2 ... Increasing the pool size improves results (Figure 7) ...

Near-optimal Adaptive Pool-based Active Learning with ...
Table 1: Theoretical Properties of Greedy Criteria for Adaptive Active Learning. Criterion. Objective ...... the 20th National Conference on Artificial Intelligence,.

Negative Results for Active Learning with Convex ... - Steve Hanneke
The hope is that the optimal solution to the surrogate risk will also have small risk ... Appearing in Proceedings of the 13th International Conference on Artificial ...

Regex-based Entity Extraction with Active Learning and ...
answer the query (button “Extract” or “Do not extract”, re- spectively). Otherwise, when the user has to describe a more complex answer, by clicking the “Edit” button the user may extend the selection boundaries of the ..... table, cell (

Negative Results for Active Learning with Convex ... - Steve Hanneke
This is the primary tool that has given ... Though negative in nature, we stress that these results should be ... responds to testing whether sign(f(x)) = y, and the risk becomes ..... active learning with applications to text classification. Journal

Microsoft Learning Experiences - GitHub
Performance for SQL Based Applications. Then, if you have not already done so, ... In the Save As dialog box, save the file as plan1.sqlplan on your desktop. 6.

Microsoft Learning Experiences - GitHub
A Windows, Linux, or Mac OS X computer. • Azure Storage Explorer. • The lab files for this course. • A Spark 2.0 HDInsight cluster. Note: If you have not already ...

Microsoft Learning Experiences - GitHub
Start Microsoft SQL Server Management Studio and connect to your database instance. 2. Click New Query, select the AdventureWorksLT database, type the ...

Microsoft Learning Experiences - GitHub
performed by writing code to manipulate data in R or Python, or by using some of the built-in modules ... https://cran.r-project.org/web/packages/dplyr/dplyr.pdf. ... You can also import custom R libraries that you have uploaded to Azure ML as R.

Microsoft Learning Experiences - GitHub
Developing SQL Databases. Lab 4 – Creating Indexes. Overview. A table named Opportunity has recently been added to the DirectMarketing schema within the database, but it has no constraints in place. In this lab, you will implement the required cons