Periodic Reporting for period 4 - DisDyn (Distributed and Dynamic Graph Algorithms and Complexity)
Reporting period: 2021-04-01 to 2022-07-31
Graph (network) data such as social networks are, like other modern big data, sheer in volume, come from a variety of sources, and evolve at a high velocity. To deal with such data, traditionally fast algorithms, i.e. those that can quickly process static graph data on a single machine, are no longer considered sufficient. Modern algorithms need to work across different computational paradigms; for example, they must be distributed, so that they can handle graph data stored at multiple machines, and dynamic, so that they can handle rapid changes. The overall goal of this project is to develop a new generation of algorithms that work across different computation paradigms and at the same time use insights achieved through lenses provided by modern computation models to make progress on long-standing questions about graph algorithms in the field. Better algorithms would allow computing devices to do the same tasks with fewer resources such as time, energy, or hardware.
The project has been carried out by a team of 4 PhD students, 3 postdocs, and the PI. The team developed a new theory that allows us to design efficient cross-paradigm algorithms for fundamental graph problems. Some of the results answer long-standing questions from 5-6 decades ago and some others were called breakthroughs by fellow scientists and received awards from flagship conferences in the field. These results are mostly theoretical, as this is the focus of the project, and the practicality of these results is under exploration.
Besides the result, the project also trained researchers that became leaders in the field; e.g. former Ph.D. students are now faculty members at top universities.
The PI and his team members have given survey talks to fellow computer scientists and students at many research institutions and events. These talks covered some dynamic algorithms and complexity studied in this project. Below are some selected activities.
-- Summer School: The PI has lectured results from this project to 200+ researchers at 19th Max Planck Advanced Course on the Foundations of Computer Science in Saarbrücken, Germany and Bernoulli Center Summer School on Modern Trends in Combinatorial Optimization in Lausanne, Switzerland.
-- Survey: The PI gave a survey talk to 200+ researchers at Highlights of Algorithms 2018 in Amsterdam, Netherlands.
Other talks include conference talks at flagship conferences (e.g. FOCS/STOC 2017-present) and invited-only workshops (e.g. Shonan meeting (Japan), Dagstuhl workshops (Germany), and Bertinoro workshops (Italy))
Besides the result, the project also trained researchers that became leaders in the field; e.g. former Ph.D. students are now faculty members at top universities.
The PI and his team members have given survey talks to fellow computer scientists and students at many research institutions and events. These talks covered some dynamic algorithms and complexity studied in this project. Below are some selected activities.
-- Summer School: The PI has lectured results from this project to 200+ researchers at 19th Max Planck Advanced Course on the Foundations of Computer Science in Saarbrücken, Germany and Bernoulli Center Summer School on Modern Trends in Combinatorial Optimization in Lausanne, Switzerland.
-- Survey: The PI gave a survey talk to 200+ researchers at Highlights of Algorithms 2018 in Amsterdam, Netherlands.
Other talks include conference talks at flagship conferences (e.g. FOCS/STOC 2017-present) and invited-only workshops (e.g. Shonan meeting (Japan), Dagstuhl workshops (Germany), and Bertinoro workshops (Italy))
* The project has led to much more understanding of graph algorithms, far beyond what we originally expected in the proposal. This includes new distributed shortest paths and connectivity algorithms, and new dynamic shortest paths and connectivity algorithms.
* Beyond the proposal's expectation, the project has also led to more efficient algorithms in various settings. This includes faster algorithms for vertex connectivity and shortest paths, a deterministic linear programming solver, a deterministic max-flow algorithm, and connectivity and matching algorithms with efficient query complexity.
* Many of the above results made the first progress in decades (five decades in some cases).
* Some results received awards from leading conferences in the field, such as FOCS, ICALP, and SOSA.
* Beyond the proposal's expectation, the project has also led to more efficient algorithms in various settings. This includes faster algorithms for vertex connectivity and shortest paths, a deterministic linear programming solver, a deterministic max-flow algorithm, and connectivity and matching algorithms with efficient query complexity.
* Many of the above results made the first progress in decades (five decades in some cases).
* Some results received awards from leading conferences in the field, such as FOCS, ICALP, and SOSA.