• Publications (chronological).
• Code and data, made publicly available.
• Abbreviations of conference and journals names, etc.
• Back to main page.

• NEW: Tracking Changing Probabilities via Dynamic Learners, in arXiv, 2024. Continues the work on prediction games: online (lifelong) multiclass probabilistic predictors for non-stationarities.
  • code for the paper, sparse moving averages (SMAs).
  • Presentation at U Alberta, Amii, AI Seminar. (May 2024)
• NEW: An Information Theoretic Score for Learning Hierarchical Concepts, (also free to download from the Frontiers site) The paper is on how to "motivate" learners, or self-supervised learning systems, to build better (higher level) concepts! Appears in Frontiers in Computational Neuroscience (special topic on Advances in Shannon-based Communications and Computations Approaches to Understanding Information Processing in the Brain, celebrating the 75th anniversary of the publication of Shannons paper on information theory), 2023.
  • Earlier: Expedition: A System for the Unsupervised Learning of a Hierarchy of Concepts O. Madani (a revival of Prediction Games!). arXiv, 2021.
  • Slides ( in pdf | in ppt ) at the AAAI23 workshop on Systems Neuroscience Approach to General Intelligence (SynAGI) Feb 2023.
• NEW: 22 Cisco graphs now available at the Stanford SNAP (many thanks to Rok Sosic): The graphs reflect communications (TCP/UDP) among hosts (nodes) in distributed applications (edges are directed and can have other attributes). Two graphs contain reference (ground truth) groupings of nodes. A README file provides additional information, and a few PYTHON scripts are provided as well. Our paper furher describes the data and reports on experiments on 21 of the 22 graphs (the 22nd graph, also with information on node grouping, was added after the paper publication). Joint work with S. Averineni, and S. Gandham, IWSPA, 2022.
  • Now, also available at UCI ml repository.
• Text Analysis via Binomial Tails O. Madani. (slides,   slides with notes,   and recording on YouTube)     DI 2021: Document Intelligence Workshop@KDD 2021.
• Binomial Tails for Community Analysis O. Madani, T. Ngo, W. Zeng, S. Averin, S. Evuru, V. Malhotra, S. Gandham, N. Yadav. arXiv 2020 (appears in the workshop on AI Cyber Security (AICS), AAAI 2022).
• ExplainIt! - A Declarative Root-cause Analysis Engine for Time Series Data. V. Jeyakumar, O. Madani, A. Parandeh, A. Kulshreshtha, W. Zeng, N. Yadav. SIGMOD 2019. Expanded verision, and Presentation title slide slides2 slides3 slides4 slides5
• Data Driven Data Center Network Security (talk). V. Jeyakumar, O. Madani, A. Parandeh, N. Yadav, at IWSPA (ACM International Workshop on Security And Privacy Analytics). 2016.

• An Empirical Comparison of Sparse vs. Embedding Techniques on Many-Class Text Classification (slides). A. Balsubramani and O. Madani. NIPS workshop on Extreme Classification, 2013.
  • Presentation video of our earlier work, in 2010, on many-class learning and index learning, recently posted on YouTube by sfbayacm.
• On Using Nearly-Independent Feature Families for High Precision and Confidence. O. Madani, M. Georg, and D. Ross. Machine Learning Journal (MLJ), Volume 92, Issues 2-3, pages 457-477, 2013. The unformatted version is available here. See next bullet for the data!
• Dataset: The dataset associated with the above paper ( YouTube Multiview Games Data ), over 100k labeled instances with over 10 feature (sub)families (of visual, auditory, text), useful for research on multiview/multimodal learning (co-training, late/early fusion, clustering, etc), is available at the UCI machine learning repository, and also here, in multiple smaller partitions.

• On Using Nearly-Independent Feature Families for High Precision and Confidence. O. Madani, M. Georg, and D. Ross. ACML 2012.   We gave a long presentation and the slides were also used for poster presentation. The poster presentation was voted best!  
• Weakly Supervised Learning of Object Segmentations from Web-Scale Video. G. Hartmann, M. Grundmann, J. Hoffman, D. Tsai, V. Kwatra, O. Madani, S. Vijayanarasimhan, I. Essa, J. Rehg, and R. Sukthankar.   Best paper award at Workshop on Web-scale Vision and Social Media, ECCV 2012.

• Discovery of Numerous Specific Topics via Term Co-Occurrence Analysis. O. Madani, and J. Yu.   SRI Technical Report, March 2011 (shorter version appears in CIKM 2010).  
• Induction of Composite Features via Grouping and Composition For Image Classification. O. Madani and B. Burns.   The (Snowbird) Learning Workshop.

• Discovery of Numerous Specific Topics via Term Co-Occurrence Analysis. (see expanded paper for improved accuracies and extensive experiments) O. Madani and J. Yu.   CIKM, 2010.  
• Discounted Deterministic Markov Decision Processes and Discounted All-Pairs Shortest Paths. O. Madani, M. Thorup, and U. Zwick.   Transactions on Algorithms, 6(32), March 2010.  
• Large-Scale Many-Class Prediction via Flat Techniques. O. Madani and J. Huang.   Workshop on Large-Scale Hierarchical Text Classification at ECIR (associated with the PASCAL Challenge on Hierarchical Text Classification), 2010.
  We obtained top ranks, number one in 5 of 12 evaluation tracks, from among 18 participants!  

• Large-Scale Many-Class Prediction via Flat Techniques. O. Madani and J. Huang.   PASCAL Challenge on Hierarchical Text Classification (short paper), Dec. 2009.  
• Learning When Concepts Abound. O. Madani, M. Connor, and W. Greiner.   Journal of Machine Learning Research (JMLR), 2009. 
  • Presentation video of our work on many-class learning and index learning, posted on YouTube by sfbayacm.
• On the Empirical Complexity of Text Classification Problems. O. Madani, H. Raghavan, and R. Jones.   SRI AI Center Technical Report, 2009. 
• Efficient Online Learning and Prediction of Users' Desktop Actions. O. Madani, H. Bui, and E. Yeh.   IJCAI, 2009. 
  • Prediction and Discovery of Users' Desktop Behavior. AAAI Spring Symposiom on Human Behavior Modeling, 2009.
• Discounted Deterministic Markov Decision Processes and Discounted All-Pairs Shortest Paths. O. Madani, M. Thorup, and U. Zwick.   ACM-SIAM SODA, 2009.
   

• Systems Learning for Complex Pattern Problems. O. Madani.  AAAI FSS, Biologically Inspired Cognitive Architectures (BICA), 2008. 
• Error-Driven Generalist+Experts (EDGE): A Multi-stage Ensemble Framework for Text Categorization. J. Huang, O. Madani, and C. Lee Giles.  ACM CIKM, 2008. 
• On Updates that Constrain the Features' Connections During Learning. O. Madani, and J. Huang.  ACM KDD, 2008. 
• Large-Scale Many-Class Learning. O. Madani and M. Connor.  SIAM Conference on Data Mining (SDM). 2008. 

• Exploring Massive Learning via a Prediction System. O. Madani.  AAAI FSS07 (on Computational Approaches to Representation Change During Learning and Development), 2007. 
• Prediction Games in Infinitely Rich Worlds.  O. Madani.  Position paper at AAAI FSS, 2007.
  • Longer technical report on the basic idea/philosophy and various considerations/issues/challenges, 2007.
• Ranked Recall: Efficient Classification by Efficient Learning of Indices that Rank.   O. Madani, M. Connor.  Yahoo! Research Technical Report, 2007.
• Recall Systems: Efficient Learning and Use of Category Indices.   O. Madani, W. Greiner, D. Kempe, and M. Salavatipour.  AISTATS, 2007.
• When will Feature Feedback help?  Quantifying the Complexity of Classification Problems.   H. Raghavan, O. Madani, R. Jones.   IJCAI Workshop on Human in the Loop Computing, 2007.

2006

• Prediction Games in Infinitely Rich Worlds.  O. Madani. 2nd Utility Based Date Mining Workshop (UBDM) at KDD, 2006.
• Learning when Concepts Abound.   O. Madani and W. Greiner. Yahoo! Research Technical Report. May, 2006.
• Active Learning with Feedback on Features and Instances.  H. Raghavan, O. Madani and R. Jones. JMLR, 2006.
• A Large Scale Analysis of Query Logs for Assessing Personalization Opportunities.   S. Wedig and O. Madani. KDD 2006.
• Generating Query Substitutions.  R. Jones, B. Rey, O. Madani, and W. Greiner. WWW 2006.
• Naive Filterbots for Robust Cold-Start Recommendations.  S.T. Park, D. Pennock, O. Madani, N. Good, D. DeCoste.  KDD 2006.
• An Empirical Comparison of Algorithms for Aggregating Expert Predictions.  V. Dani, O. Madani, D. Pennock, S. Sanghai, and B. Galebach. UAI, 2006.

• InterActive Feature Selection.  H. Raghavan, O. Madani and R. Jones. IJCAI, 2005.
• Contextual Recommender Problems.   O. Madani and D. DeCoste. UBDM at  KDD, 2005.
• Biasing Web Search Results for Topic Familiarity.  G. Kumaran, R. Jones, and O. Madani. CIKM, 2005.

• Active Model Selection.  O. Madani, D. J. Lizotte, and R. Greiner. UAI, 2004.
• Co-Validation: Using Model Disagreement on Unlabeled Data to Validate Classification Algorithms.  O. Madani, D. Pennock, and G. Flake. Neural Information Processing Systems (NIPS), 2004.
• The Budgeted Multi-Armed Bandit Problem.   O. Madani, D. J. Lizotte, and R. Greiner.  COLT, 2004.
• Empowered Learners!  Manuscript (a summery of some extensions of classical supervised learning, such as active learning), 2004.

• Budgeted Learning of Naive-Bayes Classifiers.  D. J. Lizotte, O. Madani,  and R. Greiner. UAI, 2003.
• On the Undecidability of Probabilistic Planning and Related Stochastic Optimization Problems.   O. Madani, S. Hanks, and A. Condon. AI Journal (AIJ), 2003.
• On Policy Iteration as a Newton's Method and Polynomial Policy Iteration Algorithms .  O. Madani. AAAI, 2002.
• Polynomial Value Iteration Algorithms for Deterministic MDPs.   O. Madani, UAI, 2002.
• Performance of Lookahead Control Policies in the Face of Abstractions and Approximations.   I. Levner, V. Bulitko, O. Madani and R. Greiner. SARA, 2002.
• PhD Thesis: Complexity Results for Infinite-Horizon Markov Decision Processes (abstract in text), 2000.
• Optimal Information Gathering on the Internet with Time and Cost Constraints.   O. Etzioni, S. Hanks, T. Jiang, and O. Madani. SICOMP, 2000.
• On the Undecidability of Probabilistic Planning and Partially Observable Markov Decision Problems.   O. Madani, S. Hanks, and A. Condon. AAAI, 1999.
• Fast and Intuitive Clustering of Web Documents.   O. Zamir, O. Etzioni, O. Madani, R. M. Karp. KDD, 1997.
• Efficient Information Gathering on the Internet.   O. Etzioni, S. Hanks, T. Jiang, R. M. Karp, O. Madani, and O. Waarts, IEEE FOCS, 1996.

Abbreviations of conference names, journals, etc.

• AAAI = National Conference on Artificial Intelligence (also, Association for Advancement of AI)
• ACM = Association for Computing Machinery
• ACML = Asian Machine Learning Conference
• AIJ = Artificial Intelligence Journal
• AISTATS = Conference on Artificial Intelligence and Statistics
• AAAI FSS = AAAI Fall Symposium Series
• CIKM = Conference on Information and Knowledge Management
• COLT = Conference on Learning Theory
• ECCV = European Conference on Computer Vision
• FOCS = IEEE International Conference on Foundations of Computer Science 
• IEEE = Institute of Electrical and Electronics Engineers
• IJCAI = International Joint conference on Artificial Intelligence
• IWSPA = ACM International Workshop on Security And Privacy Analytics).
• JMLR = Journal of Machine Learning Research
• KDD or SIGKDD = ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
• MLJ = Machine Learning Journal
• NIPS = Conference on Neural Information Processing Systems
• SARA = Symposium on Abstraction, Reformulation, and Approximation
• SDM = SIAM (Society of Industrial and Applied Mathematicians) Conference on Data Mining
• SIAM = Society for Industrial and Applied Mathematics
• SICOMP = SIAM Journal on Computing
• SIGMOD = ACM Special Interest Group on Management of Data
• SODA = ACM-SIAM Simposium on Discrete Algorithms
• UAI = Conference on Uncertainty in Artificial Intelligence
• UBDM = Utility Based Data Mining (workshop)
• WWW = International World Wide Web Conference

Back to main page.