Title: Novel machine learning algorithms

k-separability, projection pursuit networks, universal learning machines, almost random projection machines, transfer learning etc.

• Participants: Włodzisław Duch, M. Grochowski, T. Maszczyk and others
• Dates: 2005 - present
• Goals:Change the goal of learning, introduce measure of complexity for non-separable problems.

Related projects: Metalearning

Some talks are here.

First k-separability papers.

• Duch, W. (2006) K-separability. | PDF file.
  Lecture Notes in Computer Science 4131, 188-197, 2006.
• Grochowski M, Duch W. (2007) Learning highly non-separable Boolean functions using Constructive Feedforward Neural Network. | Abstract.
  Springer Lecture Notes in Computer Science, Vol. 4668, 180-189, 2007.

• Duch W. (2007) Towards comprehensive foundations of computational intelligence. | Abstract.
  In: W. Duch and J. Mandziuk, Challenges for Computational Intelligence. Springer Studies in Computational Intelligence, Vol. 63, 261-316, 2007.
• Duch W. (2007) Learning data structures with inherent complex logic: neurocognitive perspective. | Abstract.
  Presented at the 6th WSEAS International Conference on Computational Intelligence, Man-Machine Systems and Cybernetics (CIMMACS '07), Tenerife, Canary Islands, Spain, Dec. 14-16, 2007, pp. 294-303
• Grochowski M, Duch W. (2008) A Comparison of Methods for Learning of Highly Non-Separable Problems. | Abstract.
  Lecture Notes in Computer Science, Vol. 5097, 566-577, 2008.
  Presented at the International Conference on Artificial Intelligence and Soft Computing (ICAISC'08), Zakopane, Poland
• Grochowski M, Duch W. (2009) Constrained Learning Vector Quantization or Relaxed k-Separability
  Lecture Notes in Computer Science Vol. 5768: 151-160
  Presented at the ICANN 2009
• Maszczyk T, Grochowski M, Duch W. (2010) Discovering Data Structures using Meta-learning, Visualization and Constructive Neural Networks
  In: Advances in Machine Learning II, Springer Studies in Computational Intelligence, Vol. 262, pp. 467-484.
  Eds. Koronacki J, Ras Z.W, Wierzchon S.T, Kacprzyk J.
  Dedicated to the Memory of Professor Ryszard S. Michalski.

• Grochowski M, Duch W. (2010) Constructive Neural Network Algorithms that Solve Highly Non-Separable Problems
  In: Constructive Neural Networks, Springer Studies in Computational Intelligence Vol. 258, pp. 49-70.
  Eds. Franco L, Elizondo D.A, Jerez, J.M.
• Grochowski M, Duch W. (2011) Fast Projection Pursuit Based on Quality of Projected Clusters.
  Lecture Notes in Computer Science Vol. 6594, pp. 89-97, 2011.
  Presented at ICANGA 2011.

Other papers:

• Grochowski M, Duch W. (2008) Projection Pursuit Constructive Neural Networks Based on Quality of Projected Clusters
  Lecture Notes in Computer Science, vol. 5164, 754-762, 2008.
  Presented at: International Conference on Artificial Neural Networks (ICANN'08), Prague, Czech Republic
• Maszczyk T, Duch W. (2008) Support Vector Machines for visualization and dimensionality reduction.
  Lecture Notes in Computer Science, vol. 5163, 346-356, 2008.
  Presented at: International Conference on Artificial Neural Networks (ICANN'08), Prague, Czech Republic
• Maszczyk T, Duch W. (2008) Comparison of Shannon, Renyi and Tsallis Entropy used in Decision Trees. | Abstract.
  Lecture Notes in Computer Science, Vol. 5097, 643-651, 2008.
  Presented at the International Conference on Artificial Intelligence and Soft Computing (ICAISC'08), Zakopane, Poland
• Duch W, Maszczyk T. (2009) Universal Learning Machines.
  C.S. Leung, M. Lee, J.H. Chan (Eds.): ICONIP2009, Part II, Lecture Notes in Computer Science Vol. 5864: 206–215.
  Presented at the ICONIP 2009, Bangkok, Thailand.

• Duch W, Maszczyk T. (2009) Almost Random Projection Machine
  Lecture Notes in Computer Science Vol. 5768: 789-798
  Presented at the ICANN 2009
• Maszczyk T, Duch W. (2010) Almost Random Projection Machine with Margin Maximization and Kernel Features.
  Lecture Notes in Computer Science Vol. 6353, pp. 40-48, 2010.
  Presented at the 20th International Conference on Artificial Neural Networks (ICANN 2010), Thesalloniki, Greece, 15-18.09.2010.
• Grudziński K, Grochowski M, and Duch W. (2010) Pruning Classification Rules with Reference Vectors Selection Methods
  Lecture Notes in Computer Science Vol. 6113, pp. 347-354, 2010.
  Presented at the International Conference on Artificial Intelligence and Soft Computing (ICAISC 2010).

• Maszczyk T, Duch W. (2010) Support Feature Machine for DNA microarray data.
  Lecture Notes in Artificial Intelligence Vol. 6086, pp. 178-186, 2010.
  Presented at: 7th International Conference on Rough Sets and Current Trends in Computing (RSCTC 2010); PPTX presentation: Support Feature Machine for DNA microarray data (PPTX 1.0 MB).
• Maszczyk T, Duch W. (2010) Support Feature Machines: Support Vectors are not enough.
  World Congress on Computational Intelligence, IEEE Press, pp. 3852-3859, 2010.
  Presented at: World Congress on Computational Intelligence (WCCI), July 2010, Barcelona, Spain.
• Maszczyk T, Duch W. 2010) Triangular Visualization
  Lecture Notes in Computer Science Vol. 6113, pp. 445-452, 2010.
  Presented at the International Conference on Artificial Intelligence and Soft Computing (ICAISC 2010).

• Duch W, Maszczyk T, Grochowski M (2011) Optimal Support Features for Meta-Learning.
  Book chapter, in: Meta-learning in Computational Intelligence. Studies in Computational Intelligence. Eds: N. Jankowski, K. Grabczewski, W. Duch, Springer 2011, pp. 317-358.
• Maszczyk T, Duch W. (2012) Recursive Similarity-Based Algorithm for Deep Learning.
  T. Huang et al. (Eds.): Lecture Notes in Computer Science 7665, pp. 390–397. Springer, Heidelberg, 2012
  Presented at ICONIP 2012, Doha, Qatar,
• Blachnik M, Duch W, Maszczyk T. (2012) Feature ranking methods used for selection of prototypes.
  A.E. Villa, W. Duch, P. Erdi, F. Masulli, G. Palm, Lecture Notes in Computer Science 7553, pp. 296–304, 2012
  Presented at the ICANN 2012 conference, Lausanne, Switzerland

• Duch W, Maszczyk T, Jankowski N. (2012) Make it cheap: learning with O(nd) complexity.
  2012 IEEE World Congress on Computational Intelligence, Brisbane, Queensland, Australia, 10-15.06.2012, IJCNN, IEEE Press (ISBN: 978-1-4673-1489-3), pp. 132-135.
• Maszczyk T, Duch W. (2012) Locally Optimized Kernels.
  Lecture Notes in Computer Science Vol. 7267, pp. 412–420, 2012.
  Presented at the International Conference on Artificial Intelligence and Soft Computing (ICAISC 2012).