{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,2]],"date-time":"2026-05-02T09:59:52Z","timestamp":1777715992025,"version":"3.51.4"},"reference-count":29,"publisher":"SAGE Publications","issue":"6","license":[{"start":{"date-parts":[[2022,5,12]],"date-time":"2022-05-12T00:00:00Z","timestamp":1652313600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/journals.sagepub.com\/page\/policies\/text-and-data-mining-license"}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["The International Journal of Robotics Research"],"published-print":{"date-parts":[[2023,5]]},"abstract":"<jats:p>End-to-end learning for planning is a promising approach for finding good robot strategies in situations where the state transition, observation, and reward functions are initially unknown. Many neural network architectures for this approach have shown positive results. Across these networks, seemingly small components have been used repeatedly in different architectures, which means improving the efficiency of these components has great potential to improve the overall performance of the network. This paper aims to improve one such component: The forward propagation module. In particular, we propose Locally Connected Interrelated Network (LCI-Net) \u2013 a novel type of locally connected layer with unshared but interrelated weights \u2013 to improve the efficiency of learning stochastic transition models for planning and propagating information via the learned transition models. LCI-Net is a small differentiable neural network module that can be plugged into various existing architectures. For evaluation purposes, we apply LCI-Net to VIN and QMDP-Net. VIN is an end-to-end neural network for solving Markov Decision Processes (MDPs) whose transition and reward functions are initially unknown, while QMDP-Net is its counterpart for the Partially Observable Markov Decision Process (POMDP) whose transition, observation, and reward functions are initially unknown. Simulation tests on benchmark problems involving 2D and 3D navigation and grasping indicate promising results: Changing only the forward propagation module alone with LCI-Net improves VIN\u2019s and QMDP-Net generalisation capability by more than 3\u00d7 and 10\u00d7, respectively.<\/jats:p>","DOI":"10.1177\/02783649221093092","type":"journal-article","created":{"date-parts":[[2022,5,18]],"date-time":"2022-05-18T18:00:37Z","timestamp":1652896837000},"page":"371-384","update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":0,"title":["Locally connected interrelated network: A forward propagation primitive"],"prefix":"10.1177","volume":"42","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-5749-3043","authenticated-orcid":false,"given":"Nicholas","family":"Collins","sequence":"first","affiliation":[{"name":"School of Information Technology and Electrical Engineering, University of Queensland, Brisbane, QLD, Australia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hanna","family":"Kurniawati","sequence":"additional","affiliation":[{"name":"School of Computing, Australian National University, Canberra, ACT, Australia"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"179","published-online":{"date-parts":[[2022,5,12]]},"reference":[{"key":"bibr1-02783649221093092","doi-asserted-by":"crossref","unstructured":"Collins N, Kurniawati H (2020) Locally-connected interrelated network: a forward propagation primitive. In: Proceedings of the Fourteenth Workshop on the Algorithmic Foundations of Robotics, Oulu, Finland, 21\u201323 June 2021.","DOI":"10.1007\/978-3-030-66723-8_8"},{"key":"bibr2-02783649221093092","first-page":"3582","volume":"33","author":"Fran\u00e7ois-Lavet V","year":"2019","journal-title":"Association for the Advancement of Artificial Intelligence"},{"key":"bibr3-02783649221093092","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr.2017.769"},{"key":"bibr4-02783649221093092","unstructured":"Haarnoja T, Ajay A, Levine S, et al. (2016) Backprop KF: learning discriminative deterministic state estimators. In: NIPS Conference, Barcelona, Spain, 5\u201310 December 2016."},{"key":"bibr5-02783649221093092","unstructured":"Hausknecht M, Stone P (2015) Deep recurrent Q-learning for partially observable MDPs. In: AAAI 2015 Fall Symposium, Arlington, VI, 12\u201314 November 2015."},{"key":"bibr6-02783649221093092","unstructured":"Howard A, Roy N (2003) The robotics data set repository (radish). Available at: http:\/\/radish.sourceforge.net\/"},{"key":"bibr7-02783649221093092","unstructured":"Jonkowski R, Brock O (2016) End-to-End Learnable Histogram Filters. In Workshop on Deep Learning for Action and Interaction at NIPS, Barcelona, Spain, 5-10 December 2016."},{"key":"bibr8-02783649221093092","doi-asserted-by":"publisher","DOI":"10.1016\/S0004-3702(98)00023-X"},{"key":"bibr9-02783649221093092","unstructured":"Karkus P, Hsu D, Lee WS (2017) QMDP-net: deep learning for planning under partial observability. In: NIPS Conference, Long Beach, CA, 4\u20139 December 2017."},{"key":"bibr10-02783649221093092","unstructured":"Karkus P, Hsu D, Lee WS (2018) Particle filter networks with application to visual localization. In: CoRL Conference, Z\u00fcrich, Switzerland, 29\u201331 October 2018."},{"key":"bibr11-02783649221093092","doi-asserted-by":"crossref","unstructured":"Karkus P, Ma X, Hsu D, et al. (2019) Differentiable algorithm networks for composable robot learning. In: Robotics: Science and Systems, Breisgau, Germany, 22\u201326 June 2019.","DOI":"10.15607\/RSS.2019.XV.039"},{"key":"bibr12-02783649221093092","unstructured":"Lever G, Stafford R, Shawe-Taylor J (2014) Learning transition dynamics in MDPs with online regression and greedy feature selection. In: Neural Information Processing Systems Conference Workshop on Autonomously Learning Robots, Montreal, Canada, 12 December 2014."},{"key":"bibr13-02783649221093092","unstructured":"Lee L, Parisotto E, Chaplot DS, et al. (2018) Gated path planning networks. In: ICML Conference, Stockholm, Sweden, 10\u201315 July 2018."},{"key":"bibr14-02783649221093092","doi-asserted-by":"publisher","DOI":"10.1016\/B978-1-55860-377-6.50052-9"},{"key":"bibr15-02783649221093092","unstructured":"Mirowski P, Pascanu R, Viola F, et al. (2016) Learning to navigate in complex environments. In: ICLR Conference, San Juan, Puerto Rico, 2\u20134 May 2016."},{"key":"bibr16-02783649221093092","doi-asserted-by":"publisher","DOI":"10.1038\/nature14236"},{"key":"bibr17-02783649221093092","unstructured":"Moerland TM, Broekens J, Jonker CM (2017) Learning multimodal transition dynamics for model-based reinforcement learning. In: European Conference on Machine Learning, Skopje, Macedonia, 18\u201322 September 2017."},{"key":"bibr18-02783649221093092","unstructured":"Oh J, Guo X, Lee H, et al. (2015) Action-conditional video prediction using deep networks in atari games. In: NIPS Conference, Montreal, Canada, 7\u201312 December 2015, pp. 2863\u20132871."},{"key":"bibr19-02783649221093092","unstructured":"Oh J, Singh S, Lee H (2017) Value prediction network. In: NIPS Conference, Long Beach, CA, 4\u20139 December 2017, pp. 6118\u20136128."},{"key":"bibr20-02783649221093092","unstructured":"Okada M, Rigazio L, Aoshima T (2017) Pathe Integral Networks: Ennd-To-End Differentiable Optimal Control. arXiv preprint arXiv:1706.09597."},{"key":"bibr21-02783649221093092","unstructured":"Olah C (2019) Neural networks, types, and functional programming. Available at: http:\/\/colah.github.io\/posts\/2015-09-NN-Types-FP\/"},{"key":"bibr22-02783649221093092","volume-title":"Markov Decision Processes: Discrete Stochastic Dynamic Programming","author":"Puterman ML","year":"2014"},{"key":"bibr23-02783649221093092","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA.2019.8793889"},{"key":"bibr24-02783649221093092","doi-asserted-by":"publisher","DOI":"10.1109\/ICPR.2016.7900026"},{"key":"bibr25-02783649221093092","volume-title":"The Optimal Control of Partially Observable Markov processes","author":"Sondik E","year":"1971"},{"key":"bibr26-02783649221093092","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i11.17174"},{"key":"bibr27-02783649221093092","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2017\/700"},{"key":"bibr28-02783649221093092","first-page":"2146","volume":"33","author":"van der Pol E","year":"2020","journal-title":"Advances in Neural Information Processing Systems"},{"key":"bibr29-02783649221093092","doi-asserted-by":"crossref","unstructured":"Wahlstr\u00f6m N, Sch\u00f6n TB, Deisenroth MP (2015) Learning deep dynamical models from image pixels. In: The 17th IFAC Symposium on System Identification (SYSID), Beijing, China, 19\u201321 October 2015.","DOI":"10.1016\/j.ifacol.2015.12.271"}],"container-title":["The International Journal of Robotics Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/02783649221093092","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.1177\/02783649221093092","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/02783649221093092","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T10:17:05Z","timestamp":1777457825000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.1177\/02783649221093092"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,5,12]]},"references-count":29,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2023,5]]}},"alternative-id":["10.1177\/02783649221093092"],"URL":"https:\/\/doi.org\/10.1177\/02783649221093092","relation":{},"ISSN":["0278-3649","1741-3176"],"issn-type":[{"value":"0278-3649","type":"print"},{"value":"1741-3176","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,5,12]]}}}