{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:FMZFCQGWNLVMWA5PAYHMSCHDRH","short_pith_number":"pith:FMZFCQGW","schema_version":"1.0","canonical_sha256":"2b325140d66aeacb03af060ec908e389c95210a7731c2a5b53ceffb8abcf532d","source":{"kind":"arxiv","id":"2108.03952","version":2},"attestation_state":"computed","paper":{"title":"Safe Deep Reinforcement Learning for Multi-Agent Systems with Continuous Action Spaces","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.RO"],"primary_cat":"cs.LG","authors_text":"Athina Nisioti, Dario Pavllo, Dimitris Gkouletsos, Jonas Kohler, Konstantinos Zisis, Ziyad Sheebaelhamd","submitted_at":"2021-08-09T11:33:46Z","abstract_excerpt":"Multi-agent control problems constitute an interesting area of application for deep reinforcement learning models with continuous action spaces. Such real-world applications, however, typically come with critical safety constraints that must not be violated. In order to ensure safety, we enhance the well-known multi-agent deep deterministic policy gradient (MADDPG) framework by adding a safety layer to the deep policy network. In particular, we extend the idea of linearizing the single-step transition dynamics, as was done for single-agent systems in Safe DDPG (Dalal et al., 2018), to multi-ag"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2108.03952","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-08-09T11:33:46Z","cross_cats_sorted":["cs.RO"],"title_canon_sha256":"ec5292ac44ad6ef5c56337862387b8c4e41ef7ee38a7a7d8fce26cb327d75ea0","abstract_canon_sha256":"3327de982f454d2534c232bf14de1b750e8e6ed7f090d5c659821b40d01061fc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:05:04.284483Z","signature_b64":"4DHmQJfeMnL3s5FFMiNFfiiMMPkExJmf/qe66Ezl/er55YsM5UQhwSeOJb6O2qwY9xJSq2cDezVDGv3iIY/MAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2b325140d66aeacb03af060ec908e389c95210a7731c2a5b53ceffb8abcf532d","last_reissued_at":"2026-07-05T03:05:04.284064Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:05:04.284064Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Safe Deep Reinforcement Learning for Multi-Agent Systems with Continuous Action Spaces","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.RO"],"primary_cat":"cs.LG","authors_text":"Athina Nisioti, Dario Pavllo, Dimitris Gkouletsos, Jonas Kohler, Konstantinos Zisis, Ziyad Sheebaelhamd","submitted_at":"2021-08-09T11:33:46Z","abstract_excerpt":"Multi-agent control problems constitute an interesting area of application for deep reinforcement learning models with continuous action spaces. Such real-world applications, however, typically come with critical safety constraints that must not be violated. In order to ensure safety, we enhance the well-known multi-agent deep deterministic policy gradient (MADDPG) framework by adding a safety layer to the deep policy network. In particular, we extend the idea of linearizing the single-step transition dynamics, as was done for single-agent systems in Safe DDPG (Dalal et al., 2018), to multi-ag"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2108.03952","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2108.03952/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2108.03952","created_at":"2026-07-05T03:05:04.284120+00:00"},{"alias_kind":"arxiv_version","alias_value":"2108.03952v2","created_at":"2026-07-05T03:05:04.284120+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2108.03952","created_at":"2026-07-05T03:05:04.284120+00:00"},{"alias_kind":"pith_short_12","alias_value":"FMZFCQGWNLVM","created_at":"2026-07-05T03:05:04.284120+00:00"},{"alias_kind":"pith_short_16","alias_value":"FMZFCQGWNLVMWA5P","created_at":"2026-07-05T03:05:04.284120+00:00"},{"alias_kind":"pith_short_8","alias_value":"FMZFCQGW","created_at":"2026-07-05T03:05:04.284120+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2510.11491","citing_title":"Constraint-Aware Reinforcement Learning via Adaptive Action Scaling","ref_index":31,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FMZFCQGWNLVMWA5PAYHMSCHDRH","json":"https://pith.science/pith/FMZFCQGWNLVMWA5PAYHMSCHDRH.json","graph_json":"https://pith.science/api/pith-number/FMZFCQGWNLVMWA5PAYHMSCHDRH/graph.json","events_json":"https://pith.science/api/pith-number/FMZFCQGWNLVMWA5PAYHMSCHDRH/events.json","paper":"https://pith.science/paper/FMZFCQGW"},"agent_actions":{"view_html":"https://pith.science/pith/FMZFCQGWNLVMWA5PAYHMSCHDRH","download_json":"https://pith.science/pith/FMZFCQGWNLVMWA5PAYHMSCHDRH.json","view_paper":"https://pith.science/paper/FMZFCQGW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2108.03952&json=true","fetch_graph":"https://pith.science/api/pith-number/FMZFCQGWNLVMWA5PAYHMSCHDRH/graph.json","fetch_events":"https://pith.science/api/pith-number/FMZFCQGWNLVMWA5PAYHMSCHDRH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FMZFCQGWNLVMWA5PAYHMSCHDRH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FMZFCQGWNLVMWA5PAYHMSCHDRH/action/storage_attestation","attest_author":"https://pith.science/pith/FMZFCQGWNLVMWA5PAYHMSCHDRH/action/author_attestation","sign_citation":"https://pith.science/pith/FMZFCQGWNLVMWA5PAYHMSCHDRH/action/citation_signature","submit_replication":"https://pith.science/pith/FMZFCQGWNLVMWA5PAYHMSCHDRH/action/replication_record"}},"created_at":"2026-07-05T03:05:04.284120+00:00","updated_at":"2026-07-05T03:05:04.284120+00:00"}