ISDL@UW

Innovations in Digital Signal Processing and Machine Learning

Interactive Systems Design Lab (ISDL). Led by Professor Les Atlas

Congratulation to ISDL Ph.D. student Eldridge Alcantera for receiving the 2018 University of Washington Excellence in Teaching Award.

Congratulations to ISDL Ph.D. students Scott Wisdom and Tommy Powers for receiving the Best student paper award from the 2017 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics for: S. Wisdom, T. Powers, J. Pitton, L. Atlas, "Deep Recurrent NMF for Speech Separation by Unfolding Iterative Thresholding," Proc. IEEE WASPAA, New Paltz, New York, October 2017.

Our lab is best known for initiating these key concepts:

Convolutional neural networks (Atlas, L., Homma, T., and Marks, R., “An Artificial Neural Network for Spatio-Temporal Bipolar Patterns: Application to Phoneme Classification,” Proc. NIPS, 1987 and L.E. Atlas, T. Homma, R.J. Marks: "A Neural Network Model for Vowel Classification," Proc. IEEE-ICASSP, Dallas, 1987.)

From Atlas, Homma, Marks, "An Artificial Neural Network for Spatio-Temporal Bipolar Patterns: Application to Phoneme Classification," NIPS 1997.

Zhao-Atlas-Marks (Cone kernel) time-frequency distribution (e.g. Zhao, Y., Atlas, L., and Marks., R., "The Use of Cone-Shaped Kernels for Generalized Time-Frequency Representations of Nonstationary Signals," IEEE Transactions on Signal Processing 38, pp. 1084-1091, 1990).

Click for more detail from Wikipedia

Active learning for big data (e.g. Cohn, D., Atlas, L., and Ladner, R., “Improving Generalization with Active Learning,” Machine Learning 15, pp. 201-221. 1994 and Atlas, L., Cohn, D., and Ladner, R., "Training Connectionist Networks with Queries and Selective Sampling," Proc. NIPS 1989.)

From http://www.ms.k.u-tokyo.ac.jp/research.html#theory-active-learning

Joint acoustic and modulation frequency analysis and filtering (e.g. Atlas, L. and Shamma, S., "Joint Acoustic and Modulation Frequency," EURASIP Journal on Advances in Signal Processing, 2003.)

Full-capacity unitary recurrent neural networks (S. Wisdom*, T. Powers*, J. R. Hershey, J. Le Roux, L. Atlas, "Full-Capacity Unitary Recurrent Neural Networks," Advances in Neural Information Processing Systems (NIPS), Barcelona, Spain, December 2016. *Equal contribution)