“Housekeeping”

• This presentation consists of slides and audio. If you are experiencing any problems/issues, please press the F5 key on your keyboard if you’re using Windows, or Command + R if you’re on a Mac, to refresh your console, or close and re-launch the presentation. You can also view the Webcast Help Guide, by clicking on the “Help” widget in the bottom dock.

• To control volume, adjust the master volume on your computer.

• At the end of the presentation, you’ll see a survey URL on the final slide. Please take a minute to click on the link and fill it out to help us improve your next webinar experience.

• You can download a PDF of these slides by clicking on the Resources widget in the bottom dock.

• This presentation is being recorded and will be available for on-demand viewing in the next few days. You will receive an automatic e-mail notification when the recording is ready.

• If you think of a question during the presentation, please type it into the Q&A box and click on the submit button. You do not need to wait until the end of the presentation to begin submitting questions. You may also use the Q&A box (and the survey at the end) to suggest topics for future webinars of interest to you.

• 1,350+ trusted technical books and videos by leading publishers including O’Reilly, Morgan Kaufmann, others

• Online courses with assessments and certification-track mentoring, member discounts on tuition at partner institutions

• Learning Webinars on big topics (Cloud Computing/Mobile Development, Cybersecurity, Big Data, Recommender Systems, SaaS, Agile, Natural Language Processing)

• ACM Tech Packs on top current computing topics: Annotated Bibliographies compiled by subject experts

• Popular video tutorials/keynotes from ACM Digital Library, A.M. Turing Centenary talks/panels

• Podcasts with industry leaders/award winners

ACM Learning Center

http://learning.acm.org

Talk Back

• Use the Facebook widget in the bottom panel to share this presentation with friends and colleagues

• Use Twitter widget to Tweet your favorite quotes from today’s presentation with hashtag #ACMWebinarNELL

• Submit questions and comments via Twitter to @acmeducation – we’re reading them!

Never Ending Language Learning Tom Mitchell

Machine Learning Department Carnegie Mellon University

Tenet 1: We’ll never really understand learning until we build machines that • learn many different things, • over years, • and become better learners over time.

Tenet 2: Natural language understanding requires a belief system A natural language understanding system should react to text by saying either: • I understand, and already knew that • I understand, and didn’t know, but accept it • I understand, and disagree because …

NELL: Never-Ending Language Learner Inputs: • initial ontology (categories and relations) • dozen examples of each ontology predicate • the web • occasional interaction with human trainers The task: • run 24x7, forever • each day:

1. extract more facts from the web to populate the ontology 2. learn to read (perform #1) better than yesterday

NELL today

Running 24x7, since January, 12, 2010

Result: • KB with > 50 million candidate beliefs, growing daily • learning to read better each day • learning to reason, as well as read • automatically extending its ontology

Globe and Mail

Stanley Cup

hockey

NHL

Toronto

CFRB Wilson

play hired

won Maple Leafs

home town

city paper

league

Sundin

Milson

writer

radio

Air Canada Centre

team stadium

Canada

city stadium

politician country

Miller

airport

member

Toskala

Pearson

Skydome

Connaught

Sunnybrook

hospital

city company

skates helmet

uses equipment

won Red

Wings

Detroit

hometown

GM

city company

competes with

Toyota

plays in

league

Prius

Corrola

created Hino

acquired

automobile

economic sector

city stadium

NELL knowledge fragment climbing

football uses

equipment

NELL Today • eg. “diabetes”, “Avandia”, “tea”, “IBM”, “love” “baseball”

“BacteriaCausesCondition” “kitchenItem” “ClothingGoesWithClothing” …

How does NELL work?

Semi-Supervised Bootstrap Learning

Paris Pittsburgh Seattle Montpelier

mayor of arg1 live in arg1

San Francisco Berlin denial

arg1 is home of traits such as arg1

it’s underconstrained!!

anxiety selfishness London

Learn which noun phrases are cities:

hard (underconstrained) semi-supervised learning problem

Key Idea 1: Coupled semi-supervised training of many functions

much easier (more constrained) semi-supervised learning problem

person

noun phrase

NP:

person

Type 1 Coupling: Co-Training, Multi-View Learning Supervised training of 1 function: Minimize:

NP:

person

Type 1 Coupling: Co-Training, Multi-View Learning Coupled training of 2 functions: Minimize:

NP:

person

Type 1 Coupling: Co-Training, Multi-View Learning [Blum & Mitchell; 98] [Dasgupta et al; 01 ] [Ganchev et al., 08] [Sridharan & Kakade, 08] [Wang & Zhou, ICML10]

team

person athlete

coach sport

NP

athlete(NP) person(NP)

athlete(NP) NOT sport(NP) NOT athlete(NP) sport(NP)

Type 2 Coupling: Multi-task, Structured Outputs [Daume, 2008] [Bakhir et al., eds. 2007] [Roth et al., 2008] [Taskar et al., 2009] [Carlson et al., 2009]

team

person

NP:

athlete coach

sport

NP text context

distribution

NP morphology

NP HTML contexts

Multi-view, Multi-Task Coupling

coachesTeam(c,t) playsForTeam(a,t) teamPlaysSport(t,s)

playsSport(a,s)

NP1 NP2

Type 3 Coupling: Learning Relations

team

coachesTeam(c,t) playsForTeam(a,t) teamPlaysSport(t,s)

playsSport(a,s)

person

NP1

athlete

coach

sport

team

person

NP2

athlete

coach

sport

playsSport(NP1,NP2) athlete(NP1), sport(NP2)

Type 3 Coupling: Argument Types

over 2500 coupled functions in NELL

NELL: Learned reading strategies Plays_Sport(arg1,arg2): arg1_was_playing_arg2 arg2_megastar_arg1 arg2_icons_arg1

arg2_player_named_arg1 arg2_prodigy_arg1 arg1_is_the_tiger_woods_of_arg2 arg2_career_of_arg1 arg2_greats_as_arg1 arg1_plays_arg2 arg2_player_is_arg1 arg2_legends_arg1 arg1_announced_his_retirement_from_arg2 arg2_operations_chief_arg1 arg2_player_like_arg1 arg2_and_golfing_personalities_including_arg1 arg2_players_like_arg1 arg2_greats_like_arg1 arg2_players_are_steffi_graf_and_arg1 arg2_great_arg1 arg2_champ_arg1 arg2_greats_such_as_arg1 arg2_professionals_such_as_arg1 arg2_hit_by_arg1 arg2_greats_arg1 arg2_icon_arg1 arg2_stars_like_arg1 arg2_pros_like_arg1 arg1_retires_from_arg2 arg2_phenom_arg1 arg2_lesson_from_arg1 arg2_architects_robert_trent_jones_and_arg1 arg2_sensation_arg1 arg2_pros_arg1 arg2_stars_venus_and_arg1 arg2_hall_of_famer_arg1 arg2_superstar_arg1 arg2_legend_arg1 arg2_legends_such_as_arg1 arg2_players_is_arg1 arg2_pro_arg1 arg2_player_was_arg1 arg2_god_arg1 arg2_idol_arg1 arg1_was_born_to_play_arg2 arg2_star_arg1 arg2_hero_arg1 arg2_players_are_arg1 arg1_retired_from_professional_arg2 arg2_legends_as_arg1 arg2_autographed_by_arg1 arg2_champion_arg1 …

Continually Learning Extractors

Initial NELL Architecture Knowledge Base (latent variables)

Text Context patterns (CPL)

HTML-URL context patterns (SEAL)

Morphology classifier

(CML)

Beliefs

Candidate Beliefs

Evidence Integrator

Human advice

If coupled learning is the key, how can we get new coupling constraints?

Key Idea 2: Discover New Coupling Constraints

• first order, probabilistic horn clause constraints:

– learned by data mining the knowledge base – connect previously uncoupled relation predicates – infer new unread beliefs – modified version of FOIL [Quinlan]

0.93 athletePlaysSport(?x,?y) athletePlaysForTeam(?x,?z) teamPlaysSport(?z,?y)

Example Learned Horn Clauses

athletePlaysSport(?x,basketball) athleteInLeague(?x,NBA)

athletePlaysSport(?x,?y) athletePlaysForTeam(?x,?z) teamPlaysSport(?z,?y)

teamPlaysInLeague(?x,NHL) teamWonTrophy(?x,Stanley_Cup)

athleteInLeague(?x,?y) athletePlaysForTeam(?x,?z), teamPlaysInLeague(?z,?y)

cityInState(?x,?y) cityCapitalOfState(?x,?y), cityInCountry(?y,USA)

newspaperInCity(?x,New_York) companyEconomicSector(?x,media) generalizations(?x,blog)

0.95

0.93

0.91

0.90

0.88 0.62*

team

coachesTeam(c,t) playsForTeam(a,t) teamPlaysSport(t,s)

playsSport(a,s)

person

NP1

athlete

coach

sport

team

person

NP2

athlete

coach

sport

Learned Probabilistic Horn Clause Rules

0.93 playsSport(?x,?y) playsForTeam(?x,?z), teamPlaysSport(?z,?y)

Key Idea 3: Automatically extend ontology

Ontology Extension (1)

Goal: • Add new relations to ontology Approach: • For each pair of categories C1, C2,

• cluster pairs of known instances, in terms of text contexts that connect them

[Mohamed et al., EMNLP 2011]

Example Discovered Relations

Category Pair Frequent Instance Pairs Text Contexts Suggested Name

MusicInstrument Musician

sitar, George Harrison tenor sax, Stan Getz

trombone, Tommy Dorsey vibes, Lionel Hampton

ARG1 master ARG2 ARG1 virtuoso ARG2 ARG1 legend ARG2 ARG2 plays ARG1

Master

Disease Disease

pinched nerve, herniated disk tennis elbow, tendonitis

blepharospasm, dystonia ARG1 is due to ARG2

ARG1 is caused by ARG2 IsDueTo

CellType Chemical

epithelial cells, surfactant neurons, serotonin

mast cells, histomine ARG1 that release ARG2

ARG2 releasing ARG1 ThatRelease

Mammals Plant

koala bears, eucalyptus sheep, grasses goats, saplings

ARG1 eat ARG2 ARG2 eating ARG1 Eat

River City

Seine, Paris Nile, Cairo

Tiber river, Rome

ARG1 in heart of ARG2 ARG1 which flows through

ARG2 InHeartOf

[Mohamed et al. EMNLP 2011]

NELL: sample of self-added relations • athleteWonAward • animalEatsFood • languageTaughtInCity • clothingMadeFromPlant • beverageServedWithFood • fishServedWithFood • athleteBeatAthlete • athleteInjuredBodyPart • arthropodFeedsOnInsect • animalEatsVegetable • plantRepresentsEmotion • foodDecreasesRiskOfDisease

• clothingGoesWithClothing • bacteriaCausesPhysCondition • buildingMadeOfMaterial • emotionAssociatedWithDisease • foodCanCauseDisease • agriculturalProductAttractsInsect • arteryArisesFromArtery • countryHasSportsFans • bakedGoodServedWithBeverage • beverageContainsProtein • animalCanDevelopDisease • beverageMadeFromBeverage

Ontology Extension (2)

Goal: • Add new subcategories Approach: • For each category C,

• train NELL to read the relation SubsetOfC: C C

[Burr Settles]

*no new software here, just add this relation to ontology

NELL: example self-discovered subcategories Animal: • Pets

– Hamsters, Ferrets, Birds, Dog, Cats, Rabbits, Snakes, Parrots, Kittens, …

• Predators – Bears, Foxes, Wolves, Coyotes,

Snakes, Racoons, Eagles, Lions, Leopards, Hawks, Humans, …

Learned reading patterns for "arg1 and other medium sized arg2" "arg1 and other jungle arg2” "arg1 and other magnificent arg2" "arg1 and other pesky arg2" "arg1 and other mammals and arg2" "arg1 and other Ice Age arg2" "arg1 or other biting arg2" "arg1 and other marsh arg2" "arg1 and other migrant arg2” "arg1 and other monogastric arg2" "arg1 and other mythical

AnimalSubset(arg1,arg2)

NELL: example self-discovered subcategories Animal: • Pets

– Hamsters, Ferrets, Birds, Dog, Cats, Rabbits, Snakes, Parrots, Kittens, …

• Predators – Bears, Foxes, Wolves, Coyotes,

Snakes, Racoons, Eagles, Lions, Leopards, Hawks, Humans, …

Chemical: • Fossil fuels

– Carbon, Natural gas, Coal, Diesel, Monoxide, Gases, …

• Gases – Helium, Carbon dioxide, Methane,

Oxygen, Propane, Ozone, Radon…

Learned reading patterns: "arg1 and other medium sized arg2" "arg1 and other jungle arg2” "arg1 and other magnificent arg2" "arg1 and other pesky arg2" "arg1 and other mammals and arg2" "arg1 and other Ice Age arg2" "arg1 or other biting arg2" "arg1 and other marsh arg2" "arg1 and other migrant arg2” "arg1 and other monogastric arg2" "arg1 and other mythical

Learned reading patterns: "arg1 and other hydrocarbon arg2” "arg1 and other aqueous arg2” "arg1 and other hazardous air arg2" "arg1 and oxygen are arg2” "arg1 and such synthetic arg2” "arg1 as a lifting arg2" "arg1 as a tracer arg2" "arg1 as the carrier arg2” "arg1 as the inert arg2" "arg1 as the primary cleaning arg2” "arg1 and other noxious arg2" "arg1 and other trace arg2" "arg1 as the reagent

Combine reading and clustering Clustering NELL’s animals: • 1 rays; dolphin; roller; jacks; squid; snapper; big fish; game fish;

whitehead; stripers; chinook; rudd; wahoo; flounder; redfish; sea bass; .. • 2 lobster; oysters; shellfish; mussels; clams; scallops; oyster; clam;

abalone; crabmeat; mussel; • 3 snakes; turtles; turtle; crocodiles; alligators; endangered species;

tortoises; gecko; iguanas; rattlesnake; vipers; rattlesnakes; anaconda; ... • 4 dog; dogs; horses; cats; costa rica; sheep; livestock; puppy;

chickens; rabbit; duck; pests; goats; poultry; geek; bulldogs; farm animals; hen; bunnies; pit bulls; fowl; felines; poodle; pit bull; alpaca; …

• 5 geese; swans; flamingos; loons; canada geese; teal; mallards; egrets; sandhill cranes; mallard; snow geese;

• 6 flock; crows; seagulls; kitties; peacocks; seagull; rook; magpies; blackbirds; warbler; kittens;

• ….

[C. Liu, 2013]

NELL Summary • Learning

– Coupled multi-task, multi-view semi-supervised training

• Inference – Data mine the KB to learn inference rules – Scalable any-time inference via random walks

• Representation – Ontology extension

• Cluster pairs of noun phrases, based on corpus statistics • Directly read that X is a subcategory of Y

– Infer millions of latent concepts from observable text

• Curriculum – learn easiest things first, build on those to “learn to learn”

Key Idea 4: Cumulative, Staged Learning

1. Classify noun phrases (NP’s) by category 2. Classify NP pairs by relation 3. Discover rules to predict new relation instances 4. Learn which NP’s (co)refer to which latent concepts 5. Discover new relations to extend ontology 6. Learn to infer relation instances via targeted random walks 7. Learn to assign temporal scope to beliefs 8. Learn to microread single sentences 9. Vision: co-train text and visual object recognition 10. Goal-driven reading: predict, then read to

corroborate/correct 11. Make NELL a conversational agent on Twitter

Learning X improves ability to learn Y

thank you and thanks to: Darpa, Google, NSF, Intel, Yahoo!, Microsoft, Fulbright follow NELL on Twitter: @CMUNELL browse/download NELL’s KB at http://rtw.ml.cmu.edu

What next for NELL? • micro-reading [Krishnamurthy, Betteridge]

• map each sentence to belief system – agree/disagree/accept

• beyond English [Hrushka]

• add computer vision [Gupta, Chen]

• scalable inference over KB [Cohen, Gardner, Talukdar]

• merge with Freebase, Yago [Wijaya, Talukdar]

• goal-driven, targeted reader [Samadi, Kisiel]

• conversational agent on Twitter [Hrushka, Ritter]

• map to brain image data on sentence reading [Rafidi]

• Questions about this webcast? learning@acm.org

• ACM Learning Webinars (including archives): http://learning.acm.org/webinar • ACM Learning Center: http://learning.acm.org

• ACM SIGART: http://www.sigart.org

ACM: The Learning Continues…