Yesterday

1. CmpE Seminar: How can (worst-case optimal) joins be so interesting? by Semih Salihoglu, University of Waterloo
   • Start time: 11:00am, Thursday, December 13th
   • End time: 12:00pm, Thursday, December 13th
   • Where: AVS Conference Room, BM

   • Abstract:

     Worst-case optimality is perhaps the weakest notion of optimality for algorithms. A recent surprising theoretical development in databases has been the realization that the traditional join algorithms, which are based on binary joins, are not even worst-case optimal. Upon this realization, several surprisingly simple join algorithms have been developed that are provably worst-case optimal. Unlike traditional algorithms, which join subsets of tables at a time, worst-case join algorithms perform the join one attribute (or column) at a time. This talk gives an overview of several lines of work that my colleagues and I have been doing on worst-case join algorithms focusing on their application to subgraph queries. I will cover work from both distributed and serial settings. In the distributed setting, worst-case optimality is a yard-stick for two costs of an algorithm: (i) the load, i.e., amount of data per machine; and (ii) the total communication. Both load and communication complexity are at a trade-off with number of rounds an algorithm runs. I will describe how to achieve worst-case optimality in total communication and the performance of this algorithm on subgraph queries. It is an open theoretical problem to design constant-round algorithms with worst-case optimal load. In the serial setting, I will describe the optimizer of a prototype graph database called Graphflow that we are building at University of Waterloo. Graphflow's optimizer for subgraph queries mixes worst-case optimal join-style column-at-a-time processing seamlessly with traditional binary joins.

     Bio: Semih Salihoglu is an Assistant Professor at University of Waterloo. His research focuses on graph databases, distributed systems for processing graphs, and algorithms and theories for distributed evaluation of database queries. He holds a PhD from Stanford University and is a recipient of the 2018 VLDB best paper award.

   • View this event in Google Calendar
2. CmpE Seminar: How can (worst-case optimal) joins be so interesting? by Semih Salihoglu, University of Waterloo
   • Start time: 11:00am, Thursday, December 13th
   • End time: 12:00pm, Thursday, December 13th
   • Where: AVS Conference Room, BM

   • Abstract:
     Worst-case optimality is perhaps the weakest notion of optimality for algorithms. A recent surprising theoretical development in databases has been the realization that the traditional join algorithms, which are based on binary joins, are not even worst-case optimal. Upon this realization, several surprisingly simple join algorithms have been developed that are provably worst-case optimal. Unlike traditional algorithms, which join subsets of tables at a time, worst-case join algorithms perform the join one attribute (or column) at a time. This talk gives an overview of several lines of work that my colleagues and I have been doing on worst-case join algorithms focusing on their application to subgraph queries. I will cover work from both distributed and serial settings. In the distributed setting, worst-case optimality is a yard-stick for two costs of an algorithm: (i) the load, i.e., amount of data per machine; and (ii) the total communication. Both load and communication complexity are at a trade-off with number of rounds an algorithm runs. I will describe how to achieve worst-case optimality in total communication and the performance of this algorithm on subgraph queries. It is an open theoretical problem to design constant-round algorithms with worst-case optimal load. In the serial setting, I will describe the optimizer of a prototype graph database called Graphflow that we are building at University of Waterloo. Graphflow's optimizer for subgraph queries mixes worst-case optimal join-style column-at-a-time processing seamlessly with traditional binary joins.

     Bio: Semih Salihoglu is an Assistant Professor at University of Waterloo. His research focuses on graph databases, distributed systems for processing graphs, and algorithms and theories for distributed evaluation of database queries. He holds a PhD from Stanford University and is a recipient of the 2018 VLDB best paper award.

   • View this event in Google Calendar

Today

1. TETAM PhD Seminars
   • Start time: 10:00am, Friday, December 14th
   • End time: 12:00pm, Friday, December 14th
   • Where: TETAM Roof Conference Hall

   • Title: Dynamic Earthquake Rupture Simulations in the Sea of Marmara
     Speaker: Yasemin Korkusuz Öztürk

     Title: Analysis of the Repeating Earthquakes in the Marmara Sea
     Speaker: Nilay Başarır Baştürk

     Title: Developing a Complete Framework for Sentiment Analysis in Turkish
     Speaker: Cem Rıfkı Aydın

     http://tetam.boun.edu.tr/sites/default/files/2018-12-14_TETAM_PhD_Seminars_vol-5.pdf
     http://tetam.boun.edu.tr/sites/default/files/tetam_phd_seminars_announcement_fall_2018.jpg

   • View this event in Google Calendar
2. CmpE Seminar: Software Engineering for Mobile Apps by Mei Nagappan
   • Start time: 01:00pm, Friday, December 14th
   • End time: 02:00pm, Friday, December 14th
   • Where: AVS Conference Room, BM

   • Abstract:

     There has been tremendous growth in the use of mobile devices over the last few years. This growth has fueled the development of millions of software applications for these mobile devices often called as 'apps'. Current estimates indicate that there are hundreds of thousands of mobile app development teams, and hundreds of millions of mobile app users, with billions of dollars at stake. Therefore, studying mobile apps has the potential for tremendous impact.

     In addition to the impact, mobile apps have three key differences to the other kinds of software that have traditionally been examined in SE research - (a) They are almost exclusively distributed through an app store, which allows for any developer to easily release their apps; (b) The users are able to give direct feedback to the developers through the review mechanism; and (c) Mobile apps are often free to download and are monetized through advertisements.

     I will structure this talk with my research contributions around the above three differences. More specifically, I will talk about when a developer should release their app, what issues bother the users the most, how developers can prioritize their testing efforts and are advertisements the best monetization strategy. I will conclude my talk with some of my ongoing projects and future opportunities and challenges that exist in this line of research.

     Bio:

     Meiyappan (Mei) Nagappan is an Assistant Professor David R. Cheriton School of Computer Science at the University of Waterloo. Previously he was an Assistant Professor at the Software Engineering department of Rochester Institute of Technology and a postdoctoral fellow in the Software Analysis and Intelligence Lab (SAIL) at Queen’s University, Canada. His research is centered around the use of large-scale Software Engineering (SE) data to address the concerns of the various stakeholders (e.g., developers, operators, and managers). He received a Ph.D. in computer science from North Carolina State University. Dr. Nagappan has published in various top SE venues such as TSE, FSE, EMSE, JSS and IEEE Software. He has also received best paper awards at the International Working Conference on Mining Software Repositories (MSR ’12, ’15). He is currently the editor of the IEEE Software Blog, Information Director of the TSE journal, and a member of the MSR steering committee. He continues to collaborate with both industrial and academic researchers from the US, Canada, Japan, Germany, Italy, and India. You can find more at mei-nagappan.com.

   • View this event in Google Calendar
3. CmpE Seminar: Software Engineering for Mobile Apps by Mei Nagappan
   • Start time: 01:00pm, Friday, December 14th
   • End time: 02:00pm, Friday, December 14th
   • Where: AVS Conference Room, BM

   • Abstract:

     There has been tremendous growth in the use of mobile devices over the last few years. This growth has fueled the development of millions of software applications for these mobile devices often called as 'apps'. Current estimates indicate that there are hundreds of thousands of mobile app development teams, and hundreds of millions of mobile app users, with billions of dollars at stake. Therefore, studying mobile apps has the potential for tremendous impact.

     In addition to the impact, mobile apps have three key differences to the other kinds of software that have traditionally been examined in SE research - (a) They are almost exclusively distributed through an app store, which allows for any developer to easily release their apps; (b) The users are able to give direct feedback to the developers through the review mechanism; and (c) Mobile apps are often free to download and are monetized through advertisements.

     I will structure this talk with my research contributions around the above three differences. More specifically, I will talk about when a developer should release their app, what issues bother the users the most, how developers can prioritize their testing efforts and are advertisements the best monetization strategy. I will conclude my talk with some of my ongoing projects and future opportunities and challenges that exist in this line of research.

     Bio:

     Meiyappan (Mei) Nagappan is an Assistant Professor David R. Cheriton School of Computer Science at the University of Waterloo. Previously he was an Assistant Professor at the Software Engineering department of Rochester Institute of Technology and a postdoctoral fellow in the Software Analysis and Intelligence Lab (SAIL) at Queen’s University, Canada. His research is centered around the use of large-scale Software Engineering (SE) data to address the concerns of the various stakeholders (e.g., developers, operators, and managers). He received a Ph.D. in computer science from North Carolina State University. Dr. Nagappan has published in various top SE venues such as TSE, FSE, EMSE, JSS and IEEE Software. He has also received best paper awards at the International Working Conference on Mining Software Repositories (MSR ’12, ’15). He is currently the editor of the IEEE Software Blog, Information Director of the TSE journal, and a member of the MSR steering committee. He continues to collaborate with both industrial and academic researchers from the US, Canada, Japan, Germany, Italy, and India. You can find more at mei-nagappan.com.

   • View this event in Google Calendar

Monday, December 17th

1. CmpE Seminar: Leveraging Deep Transfer Learning for Multi-modal Affect Recognition in the Wild by Heysem Kaya
   • Start time: 03:00pm, Monday, December 17th
   • End time: 04:00pm, Monday, December 17th
   • Where: AVS Conference Room, BM
   • View this event in Google Calendar

Tuesday, December 18th

1. CmpE 579/700 Seminar: What comes after opening your robot's box? by Baris Akgun, Koc University
   • Start time: 12:00pm, Tuesday, December 18th
   • End time: 01:00pm, Tuesday, December 18th
   • Where: AVS Conference Room, BM

   • Summary:

     Current consumer and end-user robots are mostly purpose built and only capable of doing a single or a few tasks. The most general robots that can be found are manipulators, which until recently, has been behind cages or inside research labs. With the advent of low-cost collaborative robots, bringing (relatively) general purpose robots to end-users is becoming a possibility. The advances in machine learning and artificial intelligence are a major contributor to this potential as well. However, there is a lot to be solved to get there.

     This presentation will include challenges, research and ideas about how to make this possible using learning from demonstration and robotic self-learning. The main idea will concentrate how humans are goal-oriented and the ways that we can leverage this for learning. The following scenario will be the main motivator: "You buy a robot from the tech store, bring it home, open its box, look at its app store and realize that some of the things you want are missing! What do you do?" The talk will also include a learning approach for how a robot can express its own goals kinematically for more fluent collaborative interaction

     Bio:

     Barış Akgün is an Assistant Professor at the Koç University Computer Engineering Department. Prior to joining the faculty at KU on September 2016, he was a Post-Doctoral Fellow at the Electrical And Computer Engineering Department of The University of Texas at Austin. He earned his PhD Degree in Robotics from the Georgia Institute of Technology in 2015. He received his MSc Degree from the Computer Science Department and BSc. Degree with an extra-curricular Mechatronics minor from the Mechanical Engineering Department of Middle East Technical University in 2010 and 2007 respectively. His MSc and PhD work was on artificial intelligence, robot learning and human-robot interaction.

     His research interests lie at the intersection of Human-Robot Interaction, Artificial Learning for Robotics and Intelligence. His main research is about robots that learn from people and by themselves, and experimentally verified algorithmic human-robot interaction. He is also interested in applied machine learning for predictive maintenance, wearable devices, recommendation systems and medicine.

   • View this event in Google Calendar