Program – ALT/DS 2013

• Sunday 6th October 2013, 18:30: Reception
  
  
• Monday 7th October 2013, 09:00-10:30: Tutorial 1
  (Chair: Lee Wee Sun)
  Nader Bshouty: Exact Learning from Membership Queries - Some Techniques, Results and New Directions
  
  
• Monday 7th October 2013, 11:00-12:30: Tutorial 2
  (Chair: Eyke Hüllermeir)
  Krzysztof Dembczyński: Multi-Target Prediction
  
  
• Monday 7th October 2013, 12:30-13:45: Lunch
  
  
• Monday 7th October 2013, 13:45-14:45: Invited Talk 1
  (Chair: Thomas Zeugmann)
  Nir Ailon: Learning and Optimizing with Preferences
  
  
• Monday 7th October 2013, 15:15-16:55: ALT and DS Parallel Sessions, 4 Talks
  ALT Talks Session 1 – Online Learning
  (Chair: Nader Bshouty)
  
  • 15:15-15:40: Vladimir V'yugin: Universal Algorithm for Trading in Stock Market Based on the Method of Calibration
  • 15:40-16:05: Takahiro Fujita, Kohei Hatano and Eiji Takimoto: Combinatorial Online Prediction via Metarounding
  • 16:05-16:30: Jara Uitto and Roger Wattenhofer: On Competitive Recommendations
  • 16:30-16:55: Jiazhong Nie, Wojciech Kotłowski and Manfred Warmuth: On-line PCA with Optimal Regrets

  DS Talks Session 1 – Graph and Hypergraph Structures
  (Chair: Hendrik Blockeel)
  

  • 15:15-15:40: Tomi Silander and Tze-Yun Leong:
    A Dynamic Programming Algorithm for Learning Chain Event Graphs
  • 15:40-16:05: Kunihiro Wasa, Takeaki Uno, Kouichi Hirata and Hiroki Arimura:
    Polynomial Delay and Space Discovery of Connected and Acyclic Sub-Hypergraphs in a Hypergraph
  • 16:05-16:30: Takahisa Toda:
    Fast Compression of Large-scale Hypergraphs for Solving Combinatorial Problems
  • 16:30-16:55: Ye Xu, Dan Rockmore and Adam Kleinbaum:
    Hyperlink Prediction in Hypernetworks Using Latent Social Features

• Monday 7th October 2013, 19:00: ALT DS Steering Committee Meeting
  
  
• Tuesday 8th October 2013, 09:00-10:00: Invited Talk 2
  (Chair: Frank Stephan)
  Eiji Takimoto: Efficient Algorithms for Combinatorial Online Prediction
  
  
• Tuesday 8th October 2013, 10:00-10:30: E. Mark Gold Talk
  (Chair: Sanjay Jain)
  Azadeh Khaleghi and Daniil Ryabko: Nonparametric Multiple Change Point Estimation in Highly Dependent Time Series
  
  
• Tuesday 8th October 2013, 11:00-12:15: ALT and DS Parallel Sessions, 3 Talks
  ALT Talks Session 2 – Teaching, Query and PAC Learning
  (Chair: Timo Kötzing)
  
  • 11:00-11:25: Malte Darnstädt, Thorsten Doliwa, Hans Ulrich Simon and Sandra Zilles: Order Compression Schemes
  • 11:25-11:50: Jagadish M and Anindya Sen: Learning a Bounded-Degree Tree using Separator Queries
  • 11:50-12:15: Chihiro Shibata and Ryo Yoshinaka: PAC Learning of Some Subclasses of Context-Free Grammars with Basic Distributional Properties

  DS Talks Session 2 – Supervised and Weakly Supervised Learning
  (Chair: Saso Dzeroski)
  

  • 11:00-11:25: Claudio Rebelo De Sa, Carlos Soares, Arno Knobbe, Paulo Azevedo and Alipio Mario Jorge: Multi-Interval Discretization of Continuous Attributes for Label Ranking
  • 11:25-11:50: Pan Hu, Celine Vens, Bart Verstrynge and Hendrik Blockeel: Generalizing from Example Clusters
  • 11:50-12:15: Faiza Khan Khattak and Ansaf Salleb-Aouissi: Robust Crowd Labeling using Little Expertise

• Tuesday, 8th October 2013, 12:15-13:30: Lunch
  
  
• Tuesday, 8th October 2013, 13:30-15:10: ALT and DS Parallel Sessions, 4 Talks
  ALT Talks Session 3 – Statistical Learning Theory
  (Chair: Eiji Takimoto)
  
  • 13:30-13:55: Anna Choromanska, Krzysztof Choromanski, Geetha Jagannathan and Claire Monteleoni: Differentially-Private Learning of Low Dimensional Manifolds
  • 13:55-14:20: Nguyen Viet Cuong, Lam Si Tung Ho and Vu Dinh: Generalization and Robustness of Batched Weighted Average Algorithm with V-Geometrically Ergodic Markov Data
  • 14:20-14:45: Lee-Ad Gottlieb, Aryeh Kontorovich and Robert Krauthgamer: Adaptive Metric Dimensionality Reduction
  • 14:45-15:10: Ata Kaban and Robert Durrant: Dimension-Adaptive Bounds on Compressive FLD Classification

  DS Talks Session 3 – Learning with Tree Models and Hierarchies
  (Chair: Bernhard Pfahringer)
  

  • 13:30-13:55: Darko Aleksovski, Jus Kocijan and Saso Dzeroski:
    Model Tree Ensembles for Modeling Dynamic Systems
  • 13:55-14:20: Ivica Dimitrovski, Dragi Kocev, Suzana Loskovska and Saso Dzeroski:
    Fast and Scalable Image Retrieval using Predictive Clustering Trees
  • 14:20-14:45: Daniela Stojanova, Michelangelo Ceci, Donato Malerba and Saso Dzeroski:
    Learning Hierarchical Multi-label Classication Trees from Network Data
  • 14:45-15:10: Prem Raj Adhikari and Jaakko Hollmen:
    Mixture Models from Multiresolution 0-1 data

• Tuesday 8th October 2013, 15:40-16:55: ALT and DS Parallel Sessions, 3 Talks
  ALT Talks Session 4 – Bayesian and Stochastic Learning
  (Chair: Daniil Ryabko)
  
  • 15:40-16:05: Pierre Alquier Bayesian Methods for Low-Rank Matrix Estimation: Short Survey and Theoretical Study
  • 16:05-16:30: Tor Lattimore, Marcus Hutter and Peter Sunehag: Concentration and Confidence for Discrete Bayesian Sequence Predictors
  • 16:30-16:55: Aaditya Ramdas and Aarti Singh: Active Learning and Stochastic Convex Optimization

  DS Talks Session 4 – Learning and Discovery from Data Streams
  (Chair: Philippe Lenca)
  

  • 15:40-16:05: Joao Gama, Petr Kosina and Ezilda Almeida:
    Avoiding Anomalies in Data Stream Learning
  • 16:05-16:30: Dino Ienco, Albert Bifet, Indre Zlioblaite and Bernhard Pfahringer:
    Clustering Based Active Learning for Evolving Data Stream
  • 16:30-16:55: Max Zimmermann, Eirini Ntoutsi and Myra Spiliopoulou:
    Extracting Opinionated Sub-Features from a Stream of Product Reviews

• Tuesday 8th October 2013, 17:40: Mount Faber (Faber Point) Walk-Up from Harbourfront MRT Station Exit D
  
  
• Tuesday 8th October 2013, 18:30-22:30: Banquet, Jewel Box Restaurant in Mount Faber Cable Car Station (near Peak)
  
  
• Wednesday 9th October 2013, 09:00-10:00: Invited Talk 3
  (Chair: Johannes Fürnkranz)
  Hannu (T.T.) Toivonen: Creative Computers and Data Mining
  
  
• Wednesday 9th October 2013, 10:00-10:30: Carl H. Smith Talk
  (Chair: Akihiro Yamamoto)
  Eirini Spyropoulou, Tijl De Bie and Mario Boley:
  Mining Interesting Patterns in Multi-Relational Data with N-ary Relationships
  
  
• Wednesday 9th October 2013, 11:00-12:15: ALT and DS Parallel Sessions, 3 Talks
  ALT Talks Session 5 – Inductive Inference
  (Chair: Ryo Yoshinaka)
  
  • 11:00-11:25: Ziyuan Gao, Frank Stephan and Sandra Zilles: Partial Learning of Recursively Enumerable Languages
  • 11:25-11:50: John Case and Timo Kötzing: Topological Separations in Inductive Inference
  • 11:50-12:15: Marcus Hutter, Laurent Orseau and Tor Lattimore: Universal Knowledge-Seeking Agents for Stochastic Environments

  DS Talks Session 5 – Data Mining and Event Detection
  (Chair: João Gama)
  

  • 11:00-11:25: Matic Perovsek, Anze Vavpetic, Bojan Cestnik and Nada Lavrac:
    A Wordification Approach to Relational Data Mining
  • 11:25-11:50: Eniko Szekely, Pascal Poncelet, Florent Masseglia, Maguelonne Teisseire and Renaud Cezar:
    A Density-Based Backward Approach to Isolate Rare Events in Large-Scale Applications
  • 11:50-12:15: Kazumi Saito, Masahiro Kimura, Kouzou Ohara and Hiroshi Motoda:
    Identifying Super-Mediators of Information Diffusion in Social Networks

• Wednesday 9th October 2013, 12:15-14:20: Lunch and Business Meeting
  
  
• Wednesday 9th October 2013, 14:20-14:50: ALT and DS Parallel Sessions, 2 Talks
  ALT Talks Session 6 – Semisupervised and Unsupervised Learning
  (Chair: John Case)
  
  • 14:20-14:45: Daniil Ryabko: Unsupervised Model-Free Representation Learning
  • 14:45-15:10: Anna Choromanska, Tony Jebara, Hyungtae Kim, Mahesh Mohan and Claire Monteleoni: Fast Spectral Clustering via the Nyström Method

  DS Talks Session 6 – Scientific Discovery
  (Chair: Krzysztof Dembczyński)
  

  • 14:20-14:45: Jovan Tanevski, Ljupco Todorovski, Yannis Kalaidzidis and Saso Dzeroski:
    Inductive Process Modeling of Rab5-Rab7 Conversion in Endocytosis
  • 14:45-15:10: Wilfried Segretier and Martine Collard: SM2D:
    A modular knowledge discovery approach applied to hydrological forecasting

• Wednesday 9th October 2013, 15:40-16:55: ALT and DS Parallel Sessions, 3 (4) Talks
  ALT Talks Session 7 – Bandit Theory
  (Chair: Kohei Hatano)
  
  • 15:40-16:05: Po-Ling Loh and Sebastian Nowozin: Faster Hoeffding Racing: Bernstein Races via Jackknife Estimates
  • 16:05-16:30: Odalric-Ambrym Maillard: Robust Risk-Averse Stochastic Multi-Armed Bandits
  • 16:30-16:55: Gergely Neu and Gábor Bartók: An Efficient Algorithm for Learning with Semi-Bandit Feedback

  DS Talks Session 7 – Knowledge Discovery and Pattern Mining
  (Chair: Hannu Toivonen)
  

  • 15:40-16:05: Tetsushi Matsui, Takeaki Uno, Juzoh Umemori and Tsuyoshi Koide:
    A New Approach to String Pattern Mining with Approximate Match
  • 16:05-16:30: Anze Vavpetic, Petra Kralj Novak, Miha Grcar, Igor Mozetic and Nada Lavrac:
    Semantic Data Mining of Financial News Articles
  • 16:30-16:55: Pance Panov, Larisa Soldatova and Saso Dzeroski:
    OntoDM-KDD: Ontology for Representing the Knowledge Discovery Process