{"day":"01","year":"2020","publisher":"IEEE","language":[{"iso":"eng"}],"oa":1,"publication_status":"published","external_id":{"isi":["000712319503110"]},"doi":"10.1109/ICRA40945.2020.9196608","acknowledgement":"M.L. is supported in parts by the Austrian Science Fund (FWF) under grant Z211-N23 (Wittgenstein Award). R.H., and R.G. are partially supported by the Horizon-2020 ECSELProject grant No. 783163 (iDev40), and the Austrian Research Promotion Agency (FFG), Project No. 860424. R.H. and D.R. is partially supported by the Boeing Company.","page":"5446-5452","file":[{"success":1,"file_size":1070010,"content_type":"application/pdf","date_created":"2020-11-06T10:58:49Z","file_name":"2020_ICRA_Lechner.pdf","access_level":"open_access","creator":"dernst","checksum":"fccf7b986ac78046918a298cc6849a50","date_updated":"2020-11-06T10:58:49Z","relation":"main_file","file_id":"8733"}],"date_updated":"2023-08-22T10:40:15Z","conference":{"location":"Paris, France","start_date":"2020-05-31","name":"ICRA: International Conference on Robotics and Automation","end_date":"2020-08-31"},"_id":"8704","publication":"Proceedings - IEEE International Conference on Robotics and Automation","publication_identifier":{"issn":["10504729"],"isbn":["9781728173955"]},"type":"conference","month":"05","date_published":"2020-05-01T00:00:00Z","file_date_updated":"2020-11-06T10:58:49Z","date_created":"2020-10-25T23:01:19Z","oa_version":"Submitted Version","article_processing_charge":"No","user_id":"4359f0d1-fa6c-11eb-b949-802e58b17ae8","has_accepted_license":"1","status":"public","title":"Gershgorin loss stabilizes the recurrent neural network compartment of an end-to-end robot learning scheme","ddc":["000"],"isi":1,"citation":{"apa":"Lechner, M., Hasani, R., Rus, D., &#38; Grosu, R. (2020). Gershgorin loss stabilizes the recurrent neural network compartment of an end-to-end robot learning scheme. In <i>Proceedings - IEEE International Conference on Robotics and Automation</i> (pp. 5446–5452). Paris, France: IEEE. <a href=\"https://doi.org/10.1109/ICRA40945.2020.9196608\">https://doi.org/10.1109/ICRA40945.2020.9196608</a>","mla":"Lechner, Mathias, et al. “Gershgorin Loss Stabilizes the Recurrent Neural Network Compartment of an End-to-End Robot Learning Scheme.” <i>Proceedings - IEEE International Conference on Robotics and Automation</i>, IEEE, 2020, pp. 5446–52, doi:<a href=\"https://doi.org/10.1109/ICRA40945.2020.9196608\">10.1109/ICRA40945.2020.9196608</a>.","ieee":"M. Lechner, R. Hasani, D. Rus, and R. Grosu, “Gershgorin loss stabilizes the recurrent neural network compartment of an end-to-end robot learning scheme,” in <i>Proceedings - IEEE International Conference on Robotics and Automation</i>, Paris, France, 2020, pp. 5446–5452.","chicago":"Lechner, Mathias, Ramin Hasani, Daniela Rus, and Radu Grosu. “Gershgorin Loss Stabilizes the Recurrent Neural Network Compartment of an End-to-End Robot Learning Scheme.” In <i>Proceedings - IEEE International Conference on Robotics and Automation</i>, 5446–52. IEEE, 2020. <a href=\"https://doi.org/10.1109/ICRA40945.2020.9196608\">https://doi.org/10.1109/ICRA40945.2020.9196608</a>.","ama":"Lechner M, Hasani R, Rus D, Grosu R. Gershgorin loss stabilizes the recurrent neural network compartment of an end-to-end robot learning scheme. In: <i>Proceedings - IEEE International Conference on Robotics and Automation</i>. IEEE; 2020:5446-5452. doi:<a href=\"https://doi.org/10.1109/ICRA40945.2020.9196608\">10.1109/ICRA40945.2020.9196608</a>","short":"M. Lechner, R. Hasani, D. Rus, R. Grosu, in:, Proceedings - IEEE International Conference on Robotics and Automation, IEEE, 2020, pp. 5446–5452.","ista":"Lechner M, Hasani R, Rus D, Grosu R. 2020. Gershgorin loss stabilizes the recurrent neural network compartment of an end-to-end robot learning scheme. Proceedings - IEEE International Conference on Robotics and Automation. ICRA: International Conference on Robotics and Automation, ICRA, , 5446–5452."},"author":[{"first_name":"Mathias","id":"3DC22916-F248-11E8-B48F-1D18A9856A87","full_name":"Lechner, Mathias","last_name":"Lechner"},{"first_name":"Ramin","full_name":"Hasani, Ramin","last_name":"Hasani"},{"first_name":"Daniela","full_name":"Rus, Daniela","last_name":"Rus"},{"full_name":"Grosu, Radu","last_name":"Grosu","first_name":"Radu"}],"alternative_title":["ICRA"],"department":[{"_id":"ToHe"}],"scopus_import":"1","quality_controlled":"1","project":[{"grant_number":"Z211","_id":"25F42A32-B435-11E9-9278-68D0E5697425","call_identifier":"FWF","name":"The Wittgenstein Prize"}],"abstract":[{"lang":"eng","text":"Traditional robotic control suits require profound task-specific knowledge for designing, building and testing control software. The rise of Deep Learning has enabled end-to-end solutions to be learned entirely from data, requiring minimal knowledge about the application area. We design a learning scheme to train end-to-end linear dynamical systems (LDS)s by gradient descent in imitation learning robotic domains. We introduce a new regularization loss component together with a learning algorithm that improves the stability of the learned autonomous system, by forcing the eigenvalues of the internal state updates of an LDS to be negative reals. We evaluate our approach on a series of real-life and simulated robotic experiments, in comparison to linear and nonlinear Recurrent Neural Network (RNN) architectures. Our results show that our stabilizing method significantly improves test performance of LDS, enabling such linear models to match the performance of contemporary nonlinear RNN architectures. A video of the obstacle avoidance performance of our method on a mobile robot, in unseen environments, compared to other methods can be viewed at https://youtu.be/mhEsCoNao5E."}]}