JOINT PROGRAM OF ALT 2011 AND DS 2011 CONFERENCES

Tuesday, 4th of October, 2011

16:00-18:00 Registration desk open at the Dipoli congress centre, enter through the Gala Entrance on the seaside of the Dipoli congress centre to the Gala Lobby.

Wednesday, 5th of October, 2011

08:30-10:20 Tutorial 1 (ALT)
Room: Luolamies, Session Chair: Jyrki Kivinen

Sébastien Bubeck (Princeton University, USA):
Bandit Problems
10:20-10:40 Coffee break
10:40-12:30 Tutorial 2 (DS)
Room: Luolamies, Session Chair: Tapio Elomaa

Eyke Hüllermeier (Philipps-Universität Marburg, Germany) and Johannes Fürnkranz (Technische Universität Darmstadt, Germany):
Preference Learning
12:30-13:30 Lunch (Steering Committee Meeting)
13:30-14:30 Invited Speaker (ALT and DS)
Room: Luolamies, Session Chair: Esko Ukkonen

Yoshua Bengio (University of Montreal, Canada) and Olivier Delalleau:
On the Expressive Power of Deep Architectures
14:30-15:00 Coffee break
15:00-16:40 ALT and DS sessions in parallel
DS Session 1, Network Analysis
Room: Luolamies, Session Chair: Hannu Toivonen
15:00-15:25 Hiroshi Hirai, Bin-Hui Chou and Einoshin Suzuki
A Parameter-Free Method for Discovering Generalized Clusters in a Network
15:25-15:50 Jake Lussier and Nitesh V. Chawla:
Network Effects on Tweeting
15:50-16:15 Yuta Taniguchi and Daisuke Ikeda:
Graph Clustering Based on Optimization of a Macroscopic Structure of Clusters
16:15-16:40 Miha Grčar and Nada Lavrač:
A Methodology for Mining Document-Enriched Heterogeneous Information Networks
ALT Session 1, Inductive Inference
Room: Palaver, Session Chair: John Case
15:00-15:25 Timo Kötzing:
Iterative Learning from Positive Data and Counters
15:25-15:50 Sanjay Jain, Eric Martin, and Frank Stephan:
Robust Learning of Automatic Classes of Languages
15:50-16:15 Sanjay Jain, Eric Martin, and Frank Stephan:
Learning and Classifying
16:15-16:40 Michael Geilke and Sandra Zilles:
Learning Relational Patterns
16:40-17:15 Coffee break
17:15-18:30 ALT and DS sessions held in parallel
DS Session 2,Machine Learning I
Room: Luolamies, Session Chair: Jaakko Hollmén
17:15-17:40 Jan-Nikolas Sulzmann and Johannes Fürnkranz
Rule Stacking: An approach for compressing an ensemble of rule sets into a single classifier
17:40-18:05 Mitja Pugelj and Sašo Džeroski
Prediction of structured values using k nearest neighbours
18:05-18:30 Daniela Stojanova, Michelangelo Ceci, Annalisa Appice, Donato Malerba and Sašo Džeroski:
Global and Local Spatial Autocorrelation in Predictive Clustering Trees
ALT Session 2: Regression
Room: Palaver, Session Chair: Jyrki Kivinen
17:15-17:40 Sébastien Gerchinovitz and Jia Yuan Yu:
Adaptive and Optimal Online Linear Regression on L1-balls
17:40-18:05 Nina Vaits and Koby Crammer:
Re-Adapting the Regularization of Weights for Non-stationary Regression
18:45 Transportation by bus to Espoo Museum of Modern Art (EMMA) (Street address: Ahertajantie 5, Espoo), possibility to see the exhibition
20:00-21:30 Welcome party sponsored by the city of Espoo
21:00 Bus transportation to Otaniemi, Sello and Helsinki centre
21:45 Bus transportation to Otaniemi, Sello and Helsinki centre

Thursday, 6th of October, 2011

09:00-10:00 Invited Speaker 2 (DS)
Room: Luolamies, Session Chair: Jaakko Hollmén

Ming Li:
Information Distance and its Extensions
10:00-10:30 Winner of the Carl Smith Award
Room: Luolamies, Session Chair: Jaakko Hollmén

Sami Hanhijärvi (University of Helsinki, Finland):
Multiple Hypothesis Testing in Pattern Discovery
10:30-11:00 Coffee break
11:00-12:40 ALT and DS sessions held in parallel
DS Session 3: Data Mining and Pattern Discovery
Room: Luolamies, Session Chair: Geoffrey Holmes
11:00-11:25 Jeroen De Knijf, Anthony Liekens and Bart Goethals:
"Tell Me More": Finding Related Items from User Provided Feedback
11:25-11:50 Jana Schmidt and Stefan Kramer:
The Augmented Itemset Tree: A Data Structure for Online Maximum Frequent Pattern Mining
11:50-12:15 Masaya Nakahara, Shirou Maruyama, Tetsuji Kuboyama and Hiroshi Sakamoto:
Scalable Detection of Frequent Substrings by Grammar-Based Compression
12:15-12:40 Yannick Le Bras, Stéphane Lallich and Philippe Lenca:
Mining Classification Rules Without Support: an Anti-monotone Property of Jaccard Measure
ALT Session 3: Bandit Problems
Room: Palaver, Session Chair: Lev Reyzin
11:00-11:25 Sébastien Bubeck, Gilles Stoltz, and Jia Yuan Yu:
Lipschitz Bandits Without the Lipschitz Constant
11:25-11:50 Antoine Salomon and Jean-Yves Audibert:
Deviations of Stochastic Bandit Regret
11:50-12:15 Aurélien Garivier and Eric Moulines:
On Upper-confidence Bound Policies for Switching Bandit Problems
12:15-12:40 Alexandra Carpentier, Alessandro Lazaric, Mohammad Ghavamzadeh, Rémi Munos, and Peter Auer:
Upper-confidence-bound Algorithms for Active Learning in Multi-armed Bandits
12:40-13:30 Lunch
13:30-14:00 Business meeting
Room: Luolamies
14:00-15:00 Invited Speaker 3 (DS)
Room: Luolamies, Session Chair: Tapio Elomaa

Eyke Hüllermeier (Philipps-Universität Marburg, Germany) and Johannes Fürnkranz (Technische Universität Darmstadt, Germany):
Learning from Label Preferences
15:00-15:30 Coffee break
15:30-16:45 ALT and DS sessions held in parallel
DS Session 4: Contextual Data Mining
Room: Luolamies, Session Chair: Einoshin Suzuki
15:30-15:55 Masahiro Kimura, Kazumi Saito, Kouzou Ohara and Hiroshi Motoda:
Detecting Anti-majority Opinionists Using Value-weighted Mixture Voter Model
15:55-16:20 Nada Lavrač, Anže Vavpetič, Larisa Soldatova, Igor Trajkovski and Petra Kralj Novak:
Using Ontologies in Semantic Data Mining with SEGS and g-SEGS
16:20-16:45 Oleksiy Mazhelis, Indre Žliobaite and Mykola Pechenizkiy:
Context-aware Personal Route Recognition
ALT Session 4: Online Learning I
Room: Palaver, Session Chair: Gilles Stoltz
15:30-15:55 Constantinos Panagiotakopoulos and Petroula Tsampouka:
The Perceptron with Dynamic Margin
15:55-16:20 Manfred K. Warmuth and Wouter M. Koolen:
Combining Initial Segments of Lists
16:20-16:45 Eyal Gofer and Yishay Mansour:
Regret Minimization Algorithms for Pricing Lookback Options
16:45-17:15 Coffee break
17:15-18:30 ALT and DS sessions held in parallel
DS Session 5: Learning in on-line environments
Room: Luolamies, Session Chair: Philippe Lenca
17:15-17:40 Albert Bifet, Geoffrey Holmes and Bernhard Pfahringer:
MOA-TweetReader: Real-time Analysis in Twitter Streaming Data
17:40-18:05 Huma Lodhi and David Gilbert:
Bootstrapping Parameter Estimation in Dynamic Systems
18:05-18:30 Indre Žliobaite:
Controlled Permutations for Testing Adaptive Classifiers
ALT Session 5: Kernels and Margin-based Methods
Room: Palaver, Session Chair: Shai Ben-David
17:15-17:40 Xinhua Zhang, Ankan Saha, and S.V.N. Vishwanathan:
Accelerated Training of Max-margin Markov Networks with Kernels
17:40-18:05 Corinna Cortes and Mehryar Mohri:
Domain Adaptation in Regression
18:05-18:30 Daiki Suehiro, Kohei Hatano, and Eiji Takimoto:
Approximate Reduction from AUC Maximization to 1-norm Soft Margin Optimization
18:45 Transportation by bus to the banquet restaurant Kalastajatorppa, buses leave at Dipoli congress centre
19:30 Conference Banquet
22:00 Bus transportation to Otaniemi, Tapiola and Sello
23:00 Bus transportation to Otaniemi, Tapiola and Sello

Friday, 7th of October, 2011

09:00-10:00 Invited Speaker 4 (ALT)
Room: Luolamies, Session Chair: Thomas Zeugmann
Peter Auer (University of Loeben, Austria), Shiau Hong, and Chris Watkins:
Models for Autonomously Motivated Exploration in Reinforcement Learning
10:00-10:30 Winner of the E.M. Gold Award
Room: Luolamies, Session Chair: Thomas Zeugmann
Malte Darnstädt, Hans Simon, and Balázs Szörényi:
Supervised Learning and Co-training
10:30-11:00 Coffee break
11:00-12:40 ALT and DS sessions held in parallel
DS Session 6: Machine learning II
Room: Luolamies, Session Chair: Johannes Fürnkranz
11:00-11:25 Dongsheng Duan, Yuhua Li, Ruixuan Li, Zhengding Lu and Aiming Wen:
MEI: Mutual Enhanced Infinite Generative Model for Simultaneous Community and Topic Detection
11:25-11:50 Nicola Barile and Ad Feelders:
Monotone Instance Ranking with MIRA
11:50-12:15 Henna Tiensuu, Ilmari Juutilainen and Juha Röning:
Modeling the Temperature of Hot Rolled Steel Plate with Semi-supervised Learning Methods
ALT Session 6: Intelligent Agents
Room: Palaver, Session Chair: Frank Stephan
11:00-11:25 Peter Sunehag and Marcus Hutter:
Axioms for Rational Reinforcement Learning
11:25-11:50 Laurent Orseau:
Universal Knowledge-seeking Agents
11:50-12:15 Tor Lattimore and Marcus Hutter:
Asymptotically Optimal Agents
12:15-12:40 Tor Lattimore and Marcus Hutter:
Time consistent discounting
12:40-14:00 Lunch
14:00-15:00 Invited Speaker 5 (ALT)
Room: Luolamies, Session Chair: Csaba Szepesvari
Jorma Rissanen:
Optimal Estimation
15:00-15:30 Coffee break
15:30-16:45 ALT and DS sessions held in parallel
DS Session 7: Document Analysis
Room: Luolamies, Session Chair: Mika Sulkava
15:30-15:55 Mathieu Blondel, Kazuhiro Seki and Kuniaki Uehara:
Application of Semantic Kernels to Literature-Based Gene Function Annotation
15:55-16:20 Giovanni Ponti, Andrea Tagarelli and George Karypis:
A Statistical Model for Topically Segmented Documents
16:20-16:45 Christin Seifert, Eva Ulbrich and Michael Granitzer:
Word Clouds for Efficient Document Labeling
ALT Session 7: Online Learning II
Room: Palaver, Session Chair: Eiji Takimoto
15:30-15:55 Chi-Jen Lu and Wei-Fu Lu:
Making Online Decisions with Bounded Memory
15:55-16:20 Tor Lattimore, Marcus Hutter, and Vaibhav Gavane:
Universal Prediction of Selected Bits
16:20-16:45 Brendan Juba and Santosh Vempala:
Semantic Communication for Simple Goals is Equivalent to On-line Learning
16:45-17:15 Coffee break
17:15-18:30 ALT session
ALT Session 8: Other Learning Models
Room: Luolamies, Session Chair: Sanjay Jain
17:15-17:40 Ryo Yoshinaka and Anna Kasprzik:
Distributional Learning of Simple Context-free Tree Grammars
17:40-18:05 Elena Grigorescu, Lev Reyzin, and Santosh Vempala:
On Noise-tolerant Learning of Sparse Parities and Related Problems
18:05-18:30 Shalev Ben-David and Shai Ben-David:
Learning a Classifier when the Labeling is Known
18:30 End of the conference