xp36vpd -- Selected parts of data mining

Data mining aims at revealing non-trivial, hidden and ultimately applicable knowledge in large data. This course focuses on two key data mining issues: data size and their heterogeneity. When dealing with large data, it is important to resolve both the technical issues such as distributed computing or hashing and general algorithmic complexity. In this part, the course will be motivated mainly by case studies on web and social network mining. The second part will discuss approaches that merge heterogeneous prior knowledge with measured data. Bioinformatics will make the main application field here. It is assumed that students have completed at least some of the master courses on Machine Learning and Data Analysis (B4M36SAN, B4M46SMU, BE4M33SSU).

The course will take a form of reading and discussion group. Each student gives two 1 hour lectures, followed by a 30 min discussion. One of the lectures shall be DM general (MMDS book chapters, recent tutorials at major ML/DM conferences, etc.), the second one can present your research (if DM related) or a DM topic that is closely related to your research or research interests. Each student is supposed to read a review paper recommended for the topic before presentations of the other students.

Go beyond the literature, provide own insight, offer own illustrative examples, etc.

The students who do not present are supposed to read recommended reading and prepare a couple of questions before the class. The questions will be discussed during or after the talk.

Fall 2018

L	Date	Presents	Contents	Reading	Talk, other links
1	Oct 4	JK, FZ	Course overview, introduction, research interests.
2	Oct 11	JK, FZ	Preparation of the first round of presentations.
3	Oct 18	JK, FZ	Preparation of the first round of presentations.
4	Oct 26	Maria Rigaki	Adversarial ML and Generative Adversarial Networks	AML review	AML intro, GANs, Talk
5	Nov 2	Milan Němý	Deep reinforcement learning	Deep RL	AlphaGO, Talk
6	Nov 9	Jan Brabec	Non-linear dimensionality reduction – UMAP	DR review	UMAP, tSNE, Talk
7	Nov 16	Filip Paulů	On-line learning	Nonstat	ConceptDrift, OnlineDeep, Talk
8	Nov 23	Matej Uhrín	Beyond Manual Tuning of Hyperparameters	bayesOpt	HyperReview, Talk
9	Nov 30	Maria Rigaki	Gradient boosting trees	GBMTutorial	Kaggle blog, XGBoost, Talk
10	Dec 7	Milan Němý	Image segmentation with convolutional neural nets	CNNs for BioImages	DeepNetEMI, Talk
11	Dec 14	Filip Paulů	Analogous Neural Networks	HWANN	NeuromorphicComputing, Talk
12	Jan 4	Jan Brabec	Bad practices in ML	TroublingMLTrends	ClassImbalance, UsefulToKnow, Talk
13	Jan 11	Matej Uhrín	Learning for sport bets	Kelly crit	Bernoulli, Talk
14	Jan 18	JK, FZ	zkouška

References

Rajaraman, A., Leskovec, J., Ullman, J. D.: Mining of Massive Datasets, Cambridge University Press, 2011.
Free Data mining Books
Recent tutorials, major ML/DM conferences: KDD2014, DS2014, ICML16, ECML16, NIPS16
Review papers: Yang, Wu: 10 CHALLENGING PROBLEMS IN DATA MINING RESEARCH, Wu et al.: Top 10 algorithms in data mining
External seminars: ML seminars at MFF, PIS, Machine Learning Meetups, FIS KEG.

Links

Lecturers: Jiří Kléma, Filip Železný
Class schedule, meetings every Friday at 11:00 in KN:E-205. NOT as in the official schedule!
Course syllabus.

Evaluation, requirements

every student must give his talks (the principle requirement in this type of course),
attendance and active discussion at presentations of other students,
pass the exam, i.e., prove the knowledge of basic concepts presented during the course.

Table of Contents

xp36vpd -- Selected parts of data mining

Fall 2018

References

Links

Evaluation, requirements