Organization:	Michael Böhlen,  Sven Helmer and  Paolo Penna
Teaching language:	English
Level:	Advanced BSc, MSc and PhD students
Academic Year:	Spring 2020
Dates:	Monday 17.2.2020, 17.00 - 18.30, UZH, BIN 2.A.01 Saturday 25.4.2020, 9.00 - 17.00, UZH, BIN 2.A.01 Saturday 23.5.2020, 9.00 - 17.00, ETH, tba

---

Overview and objectives: The area of this year's seminar is Time Series. Students learn how to critically read and study research papers, how to summarize the contents of a paper, and how to present it in a seminar.

Teaching format: Each participant writes a self-contained report of about 10 pages and gives a 30 minutes presentation (blackboard, without a computer). Each participant has a buddy. Buddies read the report, make suggestions for improvements, and help with the presentation (e.g., dry runs). The first version of the report is due three weeks before the date of the presentation. This first version of the report and presentation will be discussed with the buddy and the teacher about two weeks before the presentation. The final versions of the report are due one week before the presentation.

Enrollment: The enrollment opens January 15, 2020 at 1pm and is handled via SDBS@OLAT. There are 18 slots in total. The first six slots of each institution are allocated first come, first served. The remaining six slots are allocated to meet constraints.

Setup and Organization: The setup of the seminar will be discussed Monday, February 17, 2020 from 17:00 until 18:30 in room BIN 2.A.01 at UZH. At the first meeting the available slots for the seminar will be distributed and papers will be assigned.

Presentations:

• Saturday April 25, BIN 2.A.01
• ​Saturday May 23, tba

Participation at all three meetings is compulsory. The assessment depends on the quality of the report, presentation, active participation during the seminar, and input as a buddy.

Useful links:

• How to give talks and read papers: link

---

Topics

1. Distance

• Dimensionality Reduction for Fast Similarity Search in Large Time Series Databases (PDF, 478 KB)
• FTW: Fast Similarity Search under the Time Warping Distance (PDF, 534 KB)
• Experiencing SAX: a Novel Symbolic Representation of Time Series (PDF, 1022 KB)
• Experimental comparison of representation methods and distance measures for time series data (PDF, 1 MB)

2. Systems

• ChronicleDB: A High-Performance Event Store (PDF, 688 KB)
• StatStream: Statistical Monitoring of Thousands of Data Streams in Real Time (PDF, 371 KB)
• Time Series Management Systems: A Survey (PDF, 4 MB)

3. Matrix Decomposition

• Faster Matrix Completion Using Randomized SVD (PDF, 4 MB)
• Memory-efficient Centroid Decomposition for Long Time Series (PDF, 617 KB)
• Global Convergence of a Grassmannian Gradient Descent Algorithm for Subspace Estimation (PDF, 283 KB)

4. Correlation

• BRAID: Stream Mining through Group Lag Correlations (PDF, 1 MB)
• Fast Approximate Correlation for Massive Time-series Data (PDF, 540 KB)
• Discovering Longest-lasting Correlation in Sequence Databases (PDF, 2 MB)

5. Pattern Matching

• DynaMMo: mining and summarization of coevolving sequences with missing values (PDF, 598 KB)
• Continuous Imputation of Missing Values in Streams of Pattern-Determining Time Series​​​​​​​ (PDF, 1 MB)
• ST-MVL: Filling Missing Values in Geo-Sensory Time Series Data (PDF, 1 MB)

6. Streaming

• Streaming PCA and Subspace Tracking: The Missing Data Case (PDF, 5 MB)
• ​​​​​​​Streaming Pattern Discovery in Multiple Time-Series (PDF, 1 MB)
• ASAP: Prioritizing Attention via Time Series Smoothing (PDF, 615 KB)​​​​​​​

 

Saturday, April 25, 2020

Topic	Presenter	Buddy	Advisor
Dimensionality Reduction for Fast Similarity Search in Large Time Series Databases (PDF, 478 KB)	Rabiya Abdullah	Monica Chiosa	Paolo Penna
FTW: Fast Similarity Search under the Time Warping Distance (PDF, 534 KB)	Mohammadamin Khashkhashi Moghaddam	Manuel Beyeler	Paolo Penna
Experiencing SAX: a Novel Symbolic Representation of Time Series (PDF, 1022 KB)	Omnia Elsaadany	Guillaume Comte	Paolo Penna
Experimental comparison of representation methods and distance measures for time series data (PDF, 1 MB)	Sven Glauser	Yuaho Yao	Michael Böhlen
ChronicleDB: A High-Performance Event Store (PDF, 688 KB)	Alexander Schindler	Guillaume Wang	Sven Helmer
StatStream: Statistical Monitoring of Thousands of Data Streams in Real Time (PDF, 371 KB)	Blagoja Trajkovski	Adrian Seiterle	Sven Helmer
Time Series Management Systems: A Survey (PDF, 4 MB)	Juan Diaz Sada	Julian Minder	Sven Helmer
Faster Matrix Completion Using Randomized SVD (PDF, 4 MB)	Yuang Cheng	Tom Wartmann	Michael Böhlen
Memory-efficient Centroid Decomposition for Long Time Series (PDF, 617 KB)	Mark Szasz	Fan Feng	Michael Böhlen
Global Convergence of a Grassmannian Gradient Descent Algorithm for Subspace Estimation (PDF, 283 KB)	-	-	-

Saturday, May 23, 2020

Topic	Presenter	Buddy	Advisor
BRAID: Stream Mining through Group Lag Correlations (PDF, 1 MB)	Monica Chiosa	Omnia Elsaadany	Sven Helmer
Fast Approximate Correlation for Massive Time-series Data (PDF, 540 KB)	Manuel Beyeler	Juan Diaz Sada	Sven Helmer
Discovering Longest-lasting Correlation in Sequence Databases (PDF, 2 MB)	Guillaume Comte	Rabiya Abdullah	Sven Helmer
DynaMMo: mining and summarization of coevolving sequences with missing values (PDF, 598 KB)	Julian Minder	Mohammadamin Khashkhashi Moghaddam	Paolo Penna
Continuous Imputation of Missing Values in Streams of Pattern-Determining Time Series​​​​​​​ (PDF, 1 MB)	Yuhao Yao	Alexander Schindler	Michael Böhlen
ST-MVL: Filling Missing Values in Geo-Sensory Time Series Data (PDF, 1 MB)	Tom Wartmann	Blagoja Trajkovski	Paolo Penna
Streaming PCA and Subspace Tracking: The Missing Data Case (PDF, 5 MB)	Guillaume Wang	Sven Glauser	Paolo Penna
Streaming Pattern Discovery in Multiple Time-Series (PDF, 1 MB)	Fan Feng	Yuang Cheng	Michael Böhlen
ASAP: Prioritizing Attention via Time Series Smoothing (PDF, 615 KB)​​​​​​​	Adrian Seiterle	Mark Szasz	Michael Böhlen