On Identifiability in Transformers Gino Brunner, Yang Liu, Damian Pascual, Oliver Richter, Massimiliano Ciaramita, Roger Wattenhofer. 8th International Conference on Learning Representations (ICLR), Online, April 2020. An early "BERTology" paper, showing limitations of interpretability.
Imperceptible Audio Communication Manuel Eichelberger, Simon Tanner, Gabriel Voirol and Roger Wattenhofer. 44th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Brighton, UK, May 2019. How to hide data in music using psychology. This paper was covered by CACM, NZZ and others.
Indoor Localization with Aircraft Signals Manuel Eichelberger, Kevin Luchsinger, Simon Tanner and Roger Wattenhofer. 15th ACM Conference on Embedded Networked Sensor Systems (SenSys), Delft, The Netherlands, November 2017. Are commercial planes better than GPS satellites because their signals are stronger and hence can still be received indoors?
Fast and Robust GPS Fix Using One Millisecond of Data Pascal Bissig, Manuel Eichelberger and Roger Wattenhofer. 16th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN), Pittsburgh, Pennsylvania, USA, April 2017.
Is 1ms of raw GPS data enough to reconstruct the position a signal was recorded, for instance to tag a photograph with its correct location?
Online Matching: Haste makes Waste! Yuval Emek, Shay Kutten and Roger Wattenhofer. 48th Annual Symposium on the Theory of Computing (STOC), Cambridge, Massachusetts, USA, June 2016.
Can we solve unsolvable online problems if we pay to delay decisions? This paper adds a completely new facet to online algorithms, inspiring a lot of excellent follow-up work.
Bitcoin Transaction Malleability and MtGox Christian Decker and Roger Wattenhofer. 19th European Symposium on Research in Computer Security (ESORICS), Wroclaw, Poland, September 2014.
We published this paper on ArXiv in March 2014, and just hours later the first media requests came in.
Achieving High Utilization with Software-Driven WAN Chi-Yao Hong, Srikanth Kandula, Ratul Mahajan, Ming Zhang, Vijay Gill, Mohan Nanduri and Roger Wattenhofer. Annual Conference of the ACM Special Interest Group on Data Communication (SIGCOMM), Hong Kong, August 2013. The "SWAN" paper, appearing in the same conference as Google's "B4" Software Defined Networking paper.
Slotted Programming for Sensor Networks Roland Flury and Roger Wattenhofer. International Conference on Information Processing in Sensor Networks (IPSN), Stockholm, Sweden, April 2010. A predecessor of the well-known Glossy protocol: There is probably no interference if two wireless nodes transmit the same information.
Tight Bounds for Clock Synchronization Christoph Lenzen, Thomas Locher and Roger Wattenhofer. Journal of the ACM, Volume 57, Number 2, New York, NY, USA, January 2010. No matter what protocol is used, and even if clock drifts and message delays are bounded by constants, the synchronization error of two neighbors in a network depends on the size of the network!
Optimal Clock Synchronization in Networks Christoph Lenzen, Philipp Sommer and Roger Wattenhofer. 7th ACM Conference on Embedded Networked Sensor Systems (SenSys), Berkeley, California, USA, November 2009. Synchronizing sensor networks with PulseSync.
Dozer: Ultra-Low Power Data Gathering in Sensor Networks Nicolas Burri, Pascal von Rickenbach and Roger Wattenhofer. International Conference on Information Processing in Sensor Networks (IPSN), Cambridge, Cambridge, Massachusetts, USA, April 2007. "This is the first paper I'd give someone working on communication in sensor nets, since it nails down how to do it right." (Matt Welsh)
The Complexity of Connectivity in Wireless Networks Thomas Moscibroda and Roger Wattenhofer. 25th Annual Joint Conference of the IEEE Computer and Communications Societies (INFOCOM), Barcelona, Spain, April 2006. The first algorithmic paper on the SINR wireless model, inspiring many more to come.
What Cannot Be Computed Locally! Fabian Kuhn, Thomas Moscibroda and Roger Wattenhofer. 23rd ACM Symposium on the Principles of Distributed Computing (PODC), St. Johns, Newfoundland, Canada, July 2004. Maximal independent set and other important problems need polylogarithmic time even in the local model, a substantial improvement over the previous lower bound, for some problems tight.