{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/backprop-kf-learning-discriminative","title":"Backprop KF: Learning Discriminative Deterministic State Estimators","arxiv_id":"1605.07148","date":"2016-05-23","proceeding":"NeurIPS 2016 12","authors":["Tuomas Haarnoja","Anurag Ajay","Sergey Levine","Pieter Abbeel"],"abstract":"Generative state estimators based on probabilistic filters and smoothers are\none of the most popular classes of state estimators for robots and autonomous\nvehicles. However, generative models have limited capacity to handle rich\nsensory observations, such as camera images, since they must model the entire\ndistribution over sensor readings. Discriminative models do not suffer from\nthis limitation, but are typically more complex to train as latent variable\nmodels for state estimation. We present an alternative approach where the\nparameters of the latent state distribution are directly optimized as a\ndeterministic computation graph, resulting in a simple and effective gradient\ndescent algorithm for training discriminative state estimators. We show that\nthis procedure can be used to train state estimators that use complex input,\nsuch as raw camera images, which must be processed using expressive nonlinear\nfunction approximators such as convolutional neural networks. Our model can be\nviewed as a type of recurrent neural network, and the connection to\nprobabilistic filtering allows us to design a network architecture that is\nparticularly well suited for state estimation. We evaluate our approach on\nsynthetic tracking task with raw image inputs and on the visual odometry task\nin the KITTI dataset. The results show significant improvement over both\nstandard generative approaches and regular recurrent neural networks.","url_abs":"http://arxiv.org/abs/1605.07148v4","url_pdf":"http://arxiv.org/pdf/1605.07148v4.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"backprop-kf-learning-discriminative","repo_url":"https://github.com/tiboat/BackpropKF_Reproduction","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"autonomous-vehicles","task_name":"Autonomous Vehicles"},{"task_slug":"state-estimation","task_name":"State Estimation"},{"task_slug":"visual-odometry","task_name":"Visual Odometry"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1605.07148","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1605.07148"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/tiboat/BackpropKF_Reproduction","reach":null}],"summary":{"ran_honours":1,"ran_draft_wrong":1,"unverified":1},"by_repo_kind":{"listed":{"samples":2,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":1,"samples":[{"code_sha256_prefix":"7337f1f5ff01dcd0","entry":"get_lr","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"7337f1f5ff01dcd0"}},{"code_sha256_prefix":"1e50303ff8e538ed","entry":"training_step","repo":"tiboat/BackpropKF_Reproduction","repo_kind":"listed","path":"Position_FF.py","file_url":"https://github.com/tiboat/BackpropKF_Reproduction/blob/HEAD/Position_FF.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"1e50303ff8e538ed"}},{"code_sha256_prefix":"b5a832ef7260e48e","entry":"evaluate","repo":"tiboat/BackpropKF_Reproduction","repo_kind":"listed","path":"Position_FF.py","file_url":"https://github.com/tiboat/BackpropKF_Reproduction/blob/HEAD/Position_FF.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b5a832ef7260e48e"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}