ReBot: a Recommendation Robot as a Fundamentally Distinct Approach towards Complexity

One of successful new programming methods in robotics is done in two steps: (1) creating a blank robot with some useful manipulators and (2) teaching the robot a given skill or process by -- figuratively -- walking it through the steps. The method has been successfully applied to dull routines like packing, carrying and/or sorting stuff, etc. Note that the method is attractive mostly because the robot can always be reset and re-taught another skill. The key feature is that humans are an integral part of the technology. In this paper, the same fundamental approach is applied to management of large sets of unstructured content where raw information is provided by folksonomies and structure is added gradually or on-demand via the proposed ReBot which requires human participation to function efficiently.

2. .
SocialRobotics
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 2/27
..
.
2
/27

3. .
SocialRoboticsis...
... whenrobots replace peopleinsocial networks
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 3/27
..
.
3
/27

4. .
.... thatwasaJOKE!(in case you have not figured it out)
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 4/27
..
.
4
/27

5. .
(thereal)SocialRobotics
02E.Berger+4"Inferring Guidance Information in Cooperative Human-Robot Tasks"HUMANOIDS (2013)
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 5/27
..
.
5
/27

6. .
SocialRoboticsinaNutshell
.
The Problem...
.
.
.
... is thatsearch spacein a robot's decision making process istoo big
.
... The Solution...
.
.
.
... is toreducesearch space viasocial interactionsbetween
robots and humans
afundamentally distinctapproach to robotics020405
social roboticsis a new area0307, but tradition is still strong06
this paper:the same basic idea applied tosoftware automation
02E.Berger+4"Inferring Guidance Information in Cooperative Human-Robot Tasks"HUMANOIDS (2013)
04A.Thomas+1"Reinforcement Learning with Human Teachers..."21st AAAI (2006)
05H.Suay+1"Effect of Human Guidance and State Space Size ...."IEEE RO-MAN (2011)
03W.Knox+2"Training a Robot via Human Feedback: A Case Study"Social Robotics (2013)
07B.Mutlu+4"Social Robotics"Springer LNAI vol.7072 (2011)
06P.Corke+0"Robotics, Vision and Control"Springer (2011)
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 6/27
..
.
6
/27

7. .
HardwarevsSoftwareRobots
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 7/27
..
.
7
/27

8. .
SoftwareRobots: theMissingTopic
01"Software Robots: The Long Tail of Automation"http://www.wired.com (2013)
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 8/27
..
.
8
/27

9. .
RobotsandReBotsasComplexity
software robots(rebots) are all aboutknowledge management
note:the same basic approach can be applied both to hardware and software

"WFSZ
DPNQMFY
TZTUFN


"MFTT
DPNQMFY
TZTUFN
5FMM
XIBU
UPEP
3PCPU
)VNBO
4FBSDI
UIFTQBDF
3PCPU
)VNBO
4FBSDI
UIFTQBDF
8IBU
TIPVME
*EP
(VJEFUISPVHI
GFFECB DL
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 9/27
..
.
9
/27

10. .
TheHardwareandSoftwareStories
1.hardware:we always know exactly whatthe jobis
software:we knowthe data, but we do not know whatknowledgeit contains
2.hardware:training is difficult becauserobots make decisions
(humans are guides)
software:training it easy sincehumans make decisions
... but robots haveto visualizethe outcome properly
3.hardware:long-term jobs are rare -- tasks are normally specific and now
software:long-term, even continuous, is the intended mode of operation
hence the ReBot -- recommendation bot is possible
recommendationscan take timeto be generated
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 10/27
..
.
10
/27

11. .
SimilaritiesandDifferences
ReBotis centered around aclassification problem08
ReBots do not have or need eyes
ReBots are dumb,humans make decision(FOFSJD
6TF
5FBDIJOH
(VJEBODF3FBTPOJOH
)VNBO
3PMF
4PDJBM
3PCPUJDT
8JEFSBOHFPG
CFIBWJPS
3FJOGPSDFE
-FBSOJOH
:&47JTJPO
SFDPHOJUJPO
(VJEFPOMZ
5IF3FCPU
QSPQPTBM "OZLJOEPG
DPOUFY U
#BZFTJB O
$MBTTJpDBUJPO
/0
/PUOFFEFE
(VJEFBOE
EFDJTJPO
NBLFS
08C.Manning+3"Introduction to Information Retrieval"Cambridge University Press (2008)
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 11/27
..
.
11
/27

12. .
ReBottheRecommendationBot
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 12/27
..
.
12
/27

13. .
ReBotversusTradition
in a way,ReBotis an idealcontext manager-- compared to existing tools09
note:multiple items is when more than one item are linked by some kind of
structure --ReBots use metromaps11.VMUJQMF
*UFNT 7JTVBM
6OEFSMZJOH
4USVDUVSF -FBSOJOH
5SBEJUJPOBM
4FBSDI
/0 /0 /0 /0
'BDFUUFE
4FBSDI
/0 /0
:&4
/0
"EWBODFE
5PPMT0UIFST
/0
:&4:&4:&4
3FCPU
UIFQSPQPTBM
:&4:&4:&4:&4
09M.Mas+0"Faceted semantic search for personalized social search"Cornell University (2011)
11K.Nesbitt+0"Getting to more abstract places using the metro map metaphor"IV Conf. (2004)
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 13/27
..
.
13
/27

14. .
ReBot: Components3FCPU
DBSFMFTT
*OQVU
)VNBO
)VNBO
\TUSVDUVSF^
QJOQPJOU
4FMFDU
#SPXTF
PSVTFPUIFSXJTF
4PNF
,OPXMFEHF
GPMLTPOPNJFT
LOPXMFEHF CBTFT
EBUBCBTFT JOEFYFT
POUPMPHJFT FUD
NFUSPNBQT
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 14/27
..
.
14
/27

15. .
ReBot: WhereisTheOriginality?
1.folksonomy-style inputis OK
folksonomy means(1)many contributors,(2)no rules,(3)careless input10
2.metromapsas the bestvisual interface12
traditional research only usesgraphs,, metromaps are very very different
new layout engines are proposed -- out of scope here13
3.recommendation enginebased on classification -- hence the ReBot
classes are based on trains and stations in metromaps
4.social scalability10
allowing for large-scale social collaboration on publically shared metromaps
10myself+0"MetroMaps versus Facets: What Exactly is the Ontological Context?"IEICE AI研究会 (November 2013)
12myself+0"...the Most Cost Effective Method in Browser-Based Visualizations..."IIEEJ Visual Computing Workshop (2013)
13myself"Rebot Project Page"http://www.github.com/maratishe/rebot (current)
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 15/27
..
.
15
/27

16. .
ReBot: SimpleImplementation
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 16/27
..
.
16
/27

17. .
ReBot: SimpleImplementation
the simple design is:
1.learning during inputvia recommendation and feedback
on input, ReBot recommends which trains to assign your paper to
you provide feedback by (possibly) picking an unexpected order
2.classification designwhich would make sure that ReBot gets
better with time
Naive Bayes classifier is used
but thedesign is unusual-- one item can belong to multiple classes
next pageshows that the engine gets better with time
3.build a greatvisual interfacebased onmetromaps and rings
metromaps:the traditional concept of metromaps
rings:simpler 3D visual aggregates of items
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 17/27
..
.
17
/27

18. .
ReBotPart1: FolksonomicalInput
(1)read the item,(2)select tags from the order recommended by ReBot,(3)let
ReBot learn from your selection -- not all assignments need learning1
2
2
3 3
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 18/27
..
.
18
/27

19. .
ReBotPart2: Recommendation0 5.2 10.4 15.6 20.8 26
Time/Learning Sequence
0
4
8
12
16
20
Rebot Order of User Selection
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 19/27
..
.
19
/27

20. .
ReBotPart3: VisualInterface
note:layouts are non-traditional13-- more suited to browser viewing
note 2:rings are simplified metromaps -- like Tokyo city with its Yamanote ring
13myself"Rebot Project Page"http://www.github.com/maratishe/rebot (current)
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 20/27
..
.
20
/27

21. .
ReBot.Papers: AWorkingExample
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 21/27
..
.
21
/27

22. .
ReBot.Papers: TheObjective
a ReBot which wouldmanage my papers
in submission, rejected, passed, ...
a verypowerful visual browsingfunction -- metromaps
ReBot wouldrecommend new stuff
similar papers in the Internet, references, etc.
recommendation to add more papers to the same train
recommendations for a more optimal metromaps for my papers
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 22/27
..
.
22
/27

23. .
ReBot.Papers: SmarterInterface
sorry, will omit small details ... but will show ademo!4VSGBDF
3FCPU
)VNBO
,OPXMFE HF#BTF UIFEBUB JUTFMG
.BJOUBJO
4FMFDUPST
USBJOT
8SJUF
5IF3BX.FUSPNBQ QBQFST
.FUSPNBQ HSPVQJOHT
$SFBUF
#SPXTF
%FQUI
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 23/27
..
.
23
/27

24. .
NextStep: CollaboratingReBots
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 24/27
..
.
24
/27

25. .
ReBotintheCloud
the screenshot:ReBot implemented asChrome Extension
withcloud storagethere is no need for web servers --APIs are
sufficient
makeslarge-scale collaborationpossible -- same for recommendations, etc.1MBDFTNFNCFST
,OPXMFEHF
JOUIFXJME
\TUSVDUVSF^
TIBSFEBDDFTT
$MPVE
4UPSBHF



7JTVBM
JOUFSGBDF
$SBXM
SFDPNNF OE
NBJOUBJO
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 25/27
..
.
25
/27

26. .
Wrapup
software robotsare proposed asrecommendation bots
share the basic idea withsocial robotics
simple designand a more complexworking exampleare presented
thesocially scalable versionis on its way
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 26/27
..
.
26
/27

27. . That’s all, thank you . ..
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 27/27
..
.
27
/27

28. .
[01](2013)
Software Robots: The Long Tail of Automation
http://www.wired.com
[02]E.Berger+4 (2013)
Inferring Guidance Information in Cooperative Human-Robot Tasks
HUMANOIDS
[03]W.Knox+2 (2013)
Training a Robot via Human Feedback: A Case Study
Social Robotics
[04]A.Thomas+1 (2006)
Reinforcement Learning with Human Teachers...
21st AAAI
[05]H.Suay+1 (2011)
Effect of Human Guidance and State Space Size ....
IEEE RO-MAN
[06]P.Corke+0 (2011)
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 27/27
..
.
27
/27

29. .
Robotics, Vision and Control
Springer
[07]B.Mutlu+4 (2011)
Social Robotics
Springer LNAI vol.7072
[08]C.Manning+3 (2008)
Introduction to Information Retrieval
Cambridge University Press
[09]M.Mas+0 (2011)
Faceted semantic search for personalized social search
Cornell University
[10]myself+0 (November 2013)
MetroMaps versus Facets: What Exactly is the Ontological Context?
IEICE AI研究会
[11]K.Nesbitt+0 (2004)
Getting to more abstract places using the metro map metaphor
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 27/27
..
.
27
/27

30. .
IV Conf.
[12]myself+0 (2013)
...the Most Cost Effective Method in Browser-Based Visualizations...
IIEEJ Visual Computing Workshop
[13]myself (current)
Rebot Project Page
http://www.github.com/maratishe/rebot
[14]myself (current)
Graphviz: Graph Visualization Software
http://www.graphviz.org/
Marat Zhanikeev -- [email protected] -- ReBoT: a Recommendation-Based Software Robot -- http://tinyurl.com/marat140213 -- 27/27
..
.
27
/27