{"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/hierarchical-foresight-self-supervised","title":"Hierarchical Foresight: Self-Supervised Learning of Long-Horizon Tasks via Visual Subgoal Generation","arxiv_id":"1909.05829","date":"2019-09-12","proceeding":"ICLR 2020 1","authors":["Suraj Nair","Chelsea Finn"],"abstract":"Video prediction models combined with planning algorithms have shown promise in enabling robots to learn to perform many vision-based tasks through only self-supervision, reaching novel goals in cluttered scenes with unseen objects. However, due to the compounding uncertainty in long horizon video prediction and poor scalability of sampling-based planning optimizers, one significant limitation of these approaches is the ability to plan over long horizons to reach distant goals. To that end, we propose a framework for subgoal generation and planning, hierarchical visual foresight (HVF), which generates subgoal images conditioned on a goal image, and uses them for planning. The subgoal images are directly optimized to decompose the task into easy to plan segments, and as a result, we observe that the method naturally identifies semantically meaningful states as subgoals. Across three out of four simulated vision-based manipulation tasks, we find that our method achieves nearly a 200% performance improvement over planning without subgoals and model-free RL approaches. Further, our experiments illustrate that our approach extends to real, cluttered visual scenes. Project page: https://sites.google.com/stanford.edu/hvf","url_abs":"https://arxiv.org/abs/1909.05829v1","url_pdf":"https://arxiv.org/pdf/1909.05829v1.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":"hierarchical-foresight-self-supervised","repo_url":"https://github.com/suraj-nair-1/google-research/tree/master/hierarchical_foresight","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"self-supervised-learning","task_name":"Self-Supervised Learning"},{"task_slug":"video-prediction","task_name":"Video Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1909.05829","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1909.05829"}},"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/suraj-nair-1/google-research/tree/master/hierarchical_foresight","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"unverified":19},"by_repo_kind":{"official":{"samples":19,"ran":0,"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":0,"samples":[{"code_sha256_prefix":"97547251dd8bf1ea","entry":"call_max_payoff","repo":"suraj-nair-1/google-research","repo_kind":"official","path":"simulation_research/tf_risk/payoffs.py","file_url":"https://github.com/suraj-nair-1/google-research/blob/HEAD/simulation_research/tf_risk/payoffs.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":"97547251dd8bf1ea"}},{"code_sha256_prefix":"af5fe347cb6b1c21","entry":"call_payoff","repo":"suraj-nair-1/google-research","repo_kind":"official","path":"simulation_research/tf_risk/payoffs.py","file_url":"https://github.com/suraj-nair-1/google-research/blob/HEAD/simulation_research/tf_risk/payoffs.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":"af5fe347cb6b1c21"}},{"code_sha256_prefix":"95c4de11b59394a7","entry":"compute_normalization_fixed_point","repo":"suraj-nair-1/google-research","repo_kind":"official","path":"bitempered_loss/loss.py","file_url":"https://github.com/suraj-nair-1/google-research/blob/HEAD/bitempered_loss/loss.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":"95c4de11b59394a7"}},{"code_sha256_prefix":"384b5cce28dc76a9","entry":"exp_t","repo":"suraj-nair-1/google-research","repo_kind":"official","path":"bitempered_loss/loss.py","file_url":"https://github.com/suraj-nair-1/google-research/blob/HEAD/bitempered_loss/loss.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":"384b5cce28dc76a9"}},{"code_sha256_prefix":"c4e259e6f90e1040","entry":"gaussian_cdf","repo":"suraj-nair-1/google-research","repo_kind":"official","path":"simulation_research/tf_risk/util.py","file_url":"https://github.com/suraj-nair-1/google-research/blob/HEAD/simulation_research/tf_risk/util.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":"c4e259e6f90e1040"}},{"code_sha256_prefix":"06faedac0b5a7f7b","entry":"half_clt_conf_interval","repo":"suraj-nair-1/google-research","repo_kind":"official","path":"simulation_research/tf_risk/util.py","file_url":"https://github.com/suraj-nair-1/google-research/blob/HEAD/simulation_research/tf_risk/util.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":"06faedac0b5a7f7b"}},{"code_sha256_prefix":"b72e0f9ec4b4e65f","entry":"hedge_derivative","repo":"suraj-nair-1/google-research","repo_kind":"official","path":"simulation_research/tf_risk/controllers.py","file_url":"https://github.com/suraj-nair-1/google-research/blob/HEAD/simulation_research/tf_risk/controllers.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":"b72e0f9ec4b4e65f"}},{"code_sha256_prefix":"cc2b7edf2f6f80c7","entry":"hist_log_euler_step","repo":"suraj-nair-1/google-research","repo_kind":"official","path":"simulation_research/tf_risk/run_replication_experiment.py","file_url":"https://github.com/suraj-nair-1/google-research/blob/HEAD/simulation_research/tf_risk/run_replication_experiment.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":"cc2b7edf2f6f80c7"}},{"code_sha256_prefix":"4dc9301a54801f0a","entry":"log_t","repo":"suraj-nair-1/google-research","repo_kind":"official","path":"bitempered_loss/loss.py","file_url":"https://github.com/suraj-nair-1/google-research/blob/HEAD/bitempered_loss/loss.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":"4dc9301a54801f0a"}},{"code_sha256_prefix":"b4b4f1f2f73b76ca","entry":"multistep_runner_tf_while","repo":"suraj-nair-1/google-research","repo_kind":"official","path":"simulation_research/tf_risk/monte_carlo_manager.py","file_url":"https://github.com/suraj-nair-1/google-research/blob/HEAD/simulation_research/tf_risk/monte_carlo_manager.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":"b4b4f1f2f73b76ca"}},{"code_sha256_prefix":"aeb9ddd5a60dd753","entry":"multistep_runner_unrolled_tf","repo":"suraj-nair-1/google-research","repo_kind":"official","path":"simulation_research/tf_risk/monte_carlo_manager.py","file_url":"https://github.com/suraj-nair-1/google-research/blob/HEAD/simulation_research/tf_risk/monte_carlo_manager.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":"aeb9ddd5a60dd753"}},{"code_sha256_prefix":"18ed7c75e6ad4c0c","entry":"non_callable_price_mc","repo":"suraj-nair-1/google-research","repo_kind":"official","path":"simulation_research/tf_risk/monte_carlo_manager.py","file_url":"https://github.com/suraj-nair-1/google-research/blob/HEAD/simulation_research/tf_risk/monte_carlo_manager.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":"18ed7c75e6ad4c0c"}},{"code_sha256_prefix":"934448b79596475e","entry":"price_derivative","repo":"suraj-nair-1/google-research","repo_kind":"official","path":"simulation_research/tf_risk/controllers.py","file_url":"https://github.com/suraj-nair-1/google-research/blob/HEAD/simulation_research/tf_risk/controllers.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":"934448b79596475e"}},{"code_sha256_prefix":"c3a241a17bb8b5ec","entry":"put_payoff","repo":"suraj-nair-1/google-research","repo_kind":"official","path":"simulation_research/tf_risk/payoffs.py","file_url":"https://github.com/suraj-nair-1/google-research/blob/HEAD/simulation_research/tf_risk/payoffs.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":"c3a241a17bb8b5ec"}},{"code_sha256_prefix":"9834d99dd9e09ef2","entry":"random_antithetic_uniform","repo":"suraj-nair-1/google-research","repo_kind":"official","path":"simulation_research/tf_risk/dynamics.py","file_url":"https://github.com/suraj-nair-1/google-research/blob/HEAD/simulation_research/tf_risk/dynamics.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":"9834d99dd9e09ef2"}},{"code_sha256_prefix":"4ece14378ed96542","entry":"random_normal","repo":"suraj-nair-1/google-research","repo_kind":"official","path":"simulation_research/tf_risk/dynamics.py","file_url":"https://github.com/suraj-nair-1/google-research/blob/HEAD/simulation_research/tf_risk/dynamics.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":"4ece14378ed96542"}},{"code_sha256_prefix":"07c5675970cb9ee8","entry":"random_uniform","repo":"suraj-nair-1/google-research","repo_kind":"official","path":"simulation_research/tf_risk/dynamics.py","file_url":"https://github.com/suraj-nair-1/google-research/blob/HEAD/simulation_research/tf_risk/dynamics.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":"07c5675970cb9ee8"}},{"code_sha256_prefix":"3f5e10b1b3013170","entry":"stddev_est","repo":"suraj-nair-1/google-research","repo_kind":"official","path":"simulation_research/tf_risk/util.py","file_url":"https://github.com/suraj-nair-1/google-research/blob/HEAD/simulation_research/tf_risk/util.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":"3f5e10b1b3013170"}},{"code_sha256_prefix":"813daf1432469b53","entry":"var_and_cvar","repo":"suraj-nair-1/google-research","repo_kind":"official","path":"simulation_research/tf_risk/risk_metrics.py","file_url":"https://github.com/suraj-nair-1/google-research/blob/HEAD/simulation_research/tf_risk/risk_metrics.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":"813daf1432469b53"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}