{
  "profile": "wrs-published-research-view-1",
  "edition": "ratings-e039106f30dec177",
  "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS",
  "entry": {
    "id": "zhu-mottaghi-et-al-image-target-siamese-navigation-2016",
    "title": "Image-target navigation with shared visual representations",
    "summary": "Instead of naming a destination, a robot can be shown a photograph of the place it should reach. Zhu and colleagues combine that target image with the current camera view in a shared Siamese network, which processes both inputs using related parameters. An actor–critic policy then chooses movement commands. Sharing representations across destinations allows experience from one route to help another. The paper tests target changes, adaptation to new simulated rooms and a small real-room robot experiment. The distinctive contribution is the goal-conditioned navigation method; the simulator resource and inherited actor–critic and visual-recognition components retain their own credit.",
    "boundary": "The main simulated setup uses four movement commands, discretized positions and fixed-size turns. New destinations can still fail, especially far from trained targets. New rooms require training scene-specific layers; the real robot uses real images and fine-tuning. This is not universal zero-shot navigation, an optimal-path guarantee or a system learned without supervised visual pretraining. The small continuous-space example still uses the same four commands.",
    "area": "Goal-conditioned visual reinforcement learning",
    "topic": "Artificial intelligence",
    "year": 2016,
    "lane": "HUMAN",
    "result": "ALGORITHM",
    "package_id": "zhu-mottaghi-et-al-image-target-siamese-navigation-2016",
    "target": {
      "record_id": "urn:uuid:0f14a6c6-2b39-49e7-a0cb-d2c772e5c340",
      "version": 1,
      "content_hash": "sha256:26ee72543daa1ed3eab74fec1c8683cae04425a0603446a476e9abfa6ac9c73b"
    },
    "humans": [
      "Yuke Zhu",
      "Roozbeh Mottaghi",
      "Eric Kolve",
      "Joseph J. Lim",
      "Abhinav Gupta",
      "Li Fei-Fei",
      "Ali Farhadi"
    ],
    "authors_complete": true,
    "human_role_note": "Source-listed human researchers and engineers for the exact method or resource. The evaluated learning system is not an independently attested scientific author. Earlier methods, funding and data-collection roles remain separate. Individual effort shares and unresolved person identities are not inferred.",
    "agents": [],
    "related": [],
    "subresults": [],
    "scientific_assessment": {
      "source_review": "SOURCE_CURATED",
      "scientific_review": "NOT_REVIEWED_BY_WRS",
      "formal_replay": "NOT_RUN",
      "provider_attestation": "NONE"
    },
    "sources": [
      {
        "id": "cur20261017-arxiv-1609-05143v1",
        "title": "Target-driven Visual Navigation in Indoor Scenes using Deep Reinforcement Learning",
        "url": "https://arxiv.org/abs/1609.05143v1",
        "date": "2016-09-16"
      }
    ],
    "points": "70000",
    "published_grade": {
      "id": "zhu-mottaghi-et-al-image-target-siamese-navigation-2016",
      "entry_id": "zhu-mottaghi-et-al-image-target-siamese-navigation-2016",
      "band": "A",
      "modifier": null,
      "target": {
        "record_id": "urn:uuid:0f14a6c6-2b39-49e7-a0cb-d2c772e5c340",
        "version": 1,
        "content_hash": "sha256:26ee72543daa1ed3eab74fec1c8683cae04425a0603446a476e9abfa6ac9c73b"
      },
      "what_changed": "Instead of naming a destination, a robot can be shown a photograph of the place it should reach. Zhu and colleagues combine that target image with the current camera view in a shared Siamese network, which processes both inputs using related parameters. An actor–critic policy then chooses movement commands. Sharing representations across destinations allows experience from one route to help another. The paper tests target changes, adaptation to new simulated rooms and a small real-room robot experiment. The distinctive contribution is the goal-conditioned navigation method; the simulator resource and inherited actor–critic and visual-recognition components retain their own credit.",
      "scope": "Instead of naming a destination, a robot can be shown a photograph of the place it should reach. Zhu and colleagues combine that target image with the current camera view in a shared Siamese network, which processes both inputs using related parameters. An actor–critic policy then chooses movement commands. Sharing representations across destinations allows experience from one route to help another. The paper tests target changes, adaptation to new simulated rooms and a small real-room robot experiment. The distinctive contribution is the goal-conditioned navigation method; the simulator resource and inherited actor–critic and visual-recognition components retain their own credit. The main simulated setup uses four movement commands, discretized positions and fixed-size turns. New destinations can still fail, especially far from trained targets. New rooms require training scene-specific layers; the real robot uses real images and fine-tuning. This is not universal zero-shot navigation, an optimal-path guarantee or a system learned without supervised visual pretraining. The small continuous-space example still uses the same four commands.",
      "status": "PROVISIONAL_EDITORIAL",
      "scientific_correctness_certified": false
    },
    "assessment_disposition": {
      "status": "ASSESSED",
      "label": "Provisionally assessed",
      "reason": "A current grade and separate reference score identify this exact contribution.",
      "target": {
        "record_id": "urn:uuid:0f14a6c6-2b39-49e7-a0cb-d2c772e5c340",
        "version": 1,
        "content_hash": "sha256:26ee72543daa1ed3eab74fec1c8683cae04425a0603446a476e9abfa6ac9c73b"
      },
      "edition": "ratings-e039106f30dec177"
    }
  },
  "version": {
    "kind": "PUBLIC_RECORD_VIEW",
    "previous": null,
    "published_at": "2026-10-23T00:00:00Z",
    "publisher": {
      "principal_id": "urn:uuid:80116931-785b-40a2-a24b-9d59db0c7e33",
      "role": "WRS_EDITORIAL_PUBLICATION"
    },
    "record": {
      "contributors": [
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Yuke Zhu",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Target-driven Visual Navigation in Indoor Scenes using Deep Reinforcement Learning",
            "url": "https://arxiv.org/abs/1609.05143v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Roozbeh Mottaghi",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Target-driven Visual Navigation in Indoor Scenes using Deep Reinforcement Learning",
            "url": "https://arxiv.org/abs/1609.05143v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Eric Kolve",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Target-driven Visual Navigation in Indoor Scenes using Deep Reinforcement Learning",
            "url": "https://arxiv.org/abs/1609.05143v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Joseph J. Lim",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Target-driven Visual Navigation in Indoor Scenes using Deep Reinforcement Learning",
            "url": "https://arxiv.org/abs/1609.05143v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Abhinav Gupta",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Target-driven Visual Navigation in Indoor Scenes using Deep Reinforcement Learning",
            "url": "https://arxiv.org/abs/1609.05143v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Li Fei-Fei",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Target-driven Visual Navigation in Indoor Scenes using Deep Reinforcement Learning",
            "url": "https://arxiv.org/abs/1609.05143v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Ali Farhadi",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Target-driven Visual Navigation in Indoor Scenes using Deep Reinforcement Learning",
            "url": "https://arxiv.org/abs/1609.05143v1"
          }
        }
      ],
      "evidence": [
        {
          "description": "Bibliographic/stated-result source; not a WRS proof verification",
          "id": "source-0",
          "kind": "source_reference",
          "source": {
            "citation": "Target-driven Visual Navigation in Indoor Scenes using Deep Reinforcement Learning",
            "url": "https://arxiv.org/abs/1609.05143v1"
          }
        }
      ],
      "origin": "HUMAN_LEGACY",
      "record_profile": "wrs-local-research-notes-0.1",
      "record_type": "research_note",
      "rrp_version": "0.1",
      "scope": {
        "description": "The main simulated setup uses four movement commands, discretized positions and fixed-size turns. New destinations can still fail, especially far from trained targets. New rooms require training scene-specific layers; the real robot uses real images and fine-tuning. This is not universal zero-shot navigation, an optimal-path guarantee or a system learned without supervised visual pretraining. The small continuous-space example still uses the same four commands.",
        "domain": "Goal-conditioned visual reinforcement learning"
      },
      "sources": [
        {
          "citation": "Target-driven Visual Navigation in Indoor Scenes using Deep Reinforcement Learning",
          "url": "https://arxiv.org/abs/1609.05143v1"
        }
      ],
      "statement": "# Image-target navigation with shared visual representations\n\nInstead of naming a destination, a robot can be shown a photograph of the place it should reach. Zhu and colleagues combine that target image with the current camera view in a shared Siamese network, which processes both inputs using related parameters. An actor–critic policy then chooses movement commands. Sharing representations across destinations allows experience from one route to help another. The paper tests target changes, adaptation to new simulated rooms and a small real-room robot experiment. The distinctive contribution is the goal-conditioned navigation method; the simulator resource and inherited actor–critic and visual-recognition components retain their own credit.\n\n## Scientific scope\n\nThe main simulated setup uses four movement commands, discretized positions and fixed-size turns. New destinations can still fail, especially far from trained targets. New rooms require training scene-specific layers; the real robot uses real images and fine-tuning. This is not universal zero-shot navigation, an optimal-path guarantee or a system learned without supervised visual pretraining. The small continuous-space example still uses the same four commands.\n\n## Reading status\n\nThis is a source-linked WRS catalogue note. A full explanatory guide and independent scientific reconstruction have not been completed for this entry. Read the identified original work for its complete narrative and arguments.\n\n## People and systems\n\nYuke Zhu; Roozbeh Mottaghi; Eric Kolve; Joseph J. Lim; Abhinav Gupta; Li Fei-Fei; Ali Farhadi\n\nSource-listed human researchers and engineers for the exact method or resource. The evaluated learning system is not an independently attested scientific author. Earlier methods, funding and data-collection roles remain separate. Individual effort shares and unresolved person identities are not inferred.\n\n\n\n## Original sources\n\n- [Target-driven Visual Navigation in Indoor Scenes using Deep Reinforcement Learning](https://arxiv.org/abs/1609.05143v1)\n",
      "summary": "Instead of naming a destination, a robot can be shown a photograph of the place it should reach. Zhu and colleagues combine that target image with the current camera view in a shared Siamese network, which processes both inputs using related parameters. An actor–critic policy then chooses movement commands. Sharing representations across destinations allows experience from one route to help another. The paper tests target changes, adaptation to new simulated rooms and a small real-room robot experiment. The distinctive contribution is the goal-conditioned navigation method; the simulator resource and inherited actor–critic and visual-recognition components retain their own credit.",
      "title": "Image-target navigation with shared visual representations"
    },
    "record_id": "urn:uuid:0f14a6c6-2b39-49e7-a0cb-d2c772e5c340",
    "version": 1,
    "source_ref": {
      "record_id": "urn:uuid:0f14a6c6-2b39-49e7-a0cb-d2c772e5c340",
      "version": 1,
      "content_hash": "sha256:26ee72543daa1ed3eab74fec1c8683cae04425a0603446a476e9abfa6ac9c73b"
    },
    "export_view": "wrs-public-editorial-view-1",
    "public_view_hash": "sha256:b8e128c947c65207c3ff22bfd8b8f58d7a219fa63a09686ef5fd929b053f6e5a"
  },
  "target": {
    "record_id": "urn:uuid:0f14a6c6-2b39-49e7-a0cb-d2c772e5c340",
    "version": 1,
    "content_hash": "sha256:26ee72543daa1ed3eab74fec1c8683cae04425a0603446a476e9abfa6ac9c73b"
  },
  "published_grade": {
    "profile": "wrs-published-grade-1",
    "edition": "2026-10-28.1",
    "id": "zhu-mottaghi-et-al-image-target-siamese-navigation-2016",
    "entry_id": "zhu-mottaghi-et-al-image-target-siamese-navigation-2016",
    "revision": 1,
    "title": "Image-target navigation with shared visual representations",
    "context": "Goal-conditioned visual reinforcement learning",
    "track": "Reusable scientific method, experimental result or historical framework",
    "band": "A",
    "modifier": null,
    "status": "PROVISIONAL_EDITORIAL",
    "assessed_on": "2026-10-23",
    "evidence_cutoff": "2026-10-23",
    "achievement_state": "SUPPORTED_SCOPED_RESULT",
    "what_changed": "Instead of naming a destination, a robot can be shown a photograph of the place it should reach. Zhu and colleagues combine that target image with the current camera view in a shared Siamese network, which processes both inputs using related parameters. An actor–critic policy then chooses movement commands. Sharing representations across destinations allows experience from one route to help another. The paper tests target changes, adaptation to new simulated rooms and a small real-room robot experiment. The distinctive contribution is the goal-conditioned navigation method; the simulator resource and inherited actor–critic and visual-recognition components retain their own credit.",
    "scope": "Instead of naming a destination, a robot can be shown a photograph of the place it should reach. Zhu and colleagues combine that target image with the current camera view in a shared Siamese network, which processes both inputs using related parameters. An actor–critic policy then chooses movement commands. Sharing representations across destinations allows experience from one route to help another. The paper tests target changes, adaptation to new simulated rooms and a small real-room robot experiment. The distinctive contribution is the goal-conditioned navigation method; the simulator resource and inherited actor–critic and visual-recognition components retain their own credit. The main simulated setup uses four movement commands, discretized positions and fixed-size turns. New destinations can still fail, especially far from trained targets. New rooms require training scene-specific layers; the real robot uses real images and fine-tuning. This is not universal zero-shot navigation, an optimal-path guarantee or a system learned without supervised visual pretraining. The small continuous-space example still uses the same four commands.",
    "evidence_summary": "The linked original works support the scoped statement and contribution attribution. WRS has not independently replayed experiments or reconstructed a full mathematical proof. The main simulated setup uses four movement commands, discretized positions and fixed-size turns. New destinations can still fail, especially far from trained targets. New rooms require training scene-specific layers; the real robot uses real images and fine-tuning. This is not universal zero-shot navigation, an optimal-path guarantee or a system learned without supervised visual pretraining. The small continuous-space example still uses the same four commands.",
    "references": [
      {
        "title": "Target-driven Visual Navigation in Indoor Scenes using Deep Reinforcement Learning",
        "url": "https://arxiv.org/abs/1609.05143v1",
        "locator": "Original scoped contribution, statement/method and evidence limits; source-specific attribution retained."
      }
    ],
    "roles": [
      {
        "who": "The named authors in the research record",
        "credit": "SCIENCE",
        "did": "Source-listed human researchers and engineers for the exact method or resource. The evaluated learning system is not an independently attested scientific author. Earlier methods, funding and data-collection roles remain separate. Individual effort shares and unresolved person identities are not inferred."
      }
    ],
    "target": {
      "record_id": "urn:uuid:0f14a6c6-2b39-49e7-a0cb-d2c772e5c340",
      "version": 1,
      "content_hash": "sha256:26ee72543daa1ed3eab74fec1c8683cae04425a0603446a476e9abfa6ac9c73b"
    },
    "scientific_correctness_certified": false,
    "independent_expert_calibration": false,
    "method": "PROPRIETARY",
    "points": "70000"
  },
  "canonical_agent_submission": false,
  "scientific_verification_awarded": false,
  "reader_guide": null,
  "reading_status": "SOURCE_LINKED_NOTE"
}