--- abstract: 'The problem of inductive learning is hard, and--despite much work--no solution is in sight, from neural networks or other AI techniques. I suggest that inductive reasoning may be grounded in sensorimotor capacity. If an artificial system to generalize in ways that we find intelligent it should be appropriately embodied. This is illustrated with a network- controlled animat that learns n-parity by representing intermediate states with its own motion. Unlike other general learning devices, such as disembodied networks, it learns from very few examples and generalizes correctly to previously unseen cases.' altloc: [] chapter: ~ commentary: ~ commref: ~ confdates: Nov. 1996 conference: AAAI '96 Fall Workshop confloc: 'Cambridge, Massachusetts' contact_email: ~ creators_id: [] creators_name: - family: Wexler given: Mark honourific: '' lineage: '' date: 1996 date_type: published datestamp: 1998-09-18 department: ~ dir: disk0/00/00/05/15 edit_lock_since: ~ edit_lock_until: ~ edit_lock_user: ~ editors_id: [] editors_name: - family: Mataric given: Maja honourific: '' lineage: '' eprint_status: archive eprintid: 515 fileinfo: /style/images/fileicons/application_postscript.png;/515/1/final.ps|/style/images/fileicons/application_pdf.png;/515/5/final.pdf full_text_status: public importid: ~ institution: ~ isbn: ~ ispublished: pub issn: ~ item_issues_comment: [] item_issues_count: 0 item_issues_description: [] item_issues_id: [] item_issues_reported_by: [] item_issues_resolved_by: [] item_issues_status: [] item_issues_timestamp: [] item_issues_type: [] keywords: 'induction, inductive learning, generalization, action, external representations, machine learning, animats, n-parity problem' lastmod: 2011-03-11 08:54:01 latitude: ~ longitude: ~ metadata_visibility: show note: ~ number: ~ pagerange: 134-138 pubdom: FALSE publication: ~ publisher: AAAI Press refereed: FALSE referencetext: ~ relation_type: [] relation_uri: [] reportno: ~ rev_number: 14 series: ~ source: ~ status_changed: 2007-09-12 16:29:44 subjects: - comp-sci-mach-learn - phil-epist succeeds: ~ suggestions: ~ sword_depositor: ~ sword_slug: ~ thesistype: ~ title: 'Embodied induction: Learning external representations' type: confpaper userid: 506 volume: ~