{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/loss-calibrated-approximate-inference-in","title":"Loss-Calibrated Approximate Inference in Bayesian Neural Networks","arxiv_id":"1805.03901","date":"2018-05-10","proceeding":null,"authors":["Adam D. Cobb","Stephen J. Roberts","Yarin Gal"],"abstract":"Current approaches in approximate inference for Bayesian neural networks\nminimise the Kullback-Leibler divergence to approximate the true posterior over\nthe weights. However, this approximation is without knowledge of the final\napplication, and therefore cannot guarantee optimal predictions for a given\ntask. To make more suitable task-specific approximations, we introduce a new\nloss-calibrated evidence lower bound for Bayesian neural networks in the\ncontext of supervised learning, informed by Bayesian decision theory. By\nintroducing a lower bound that depends on a utility function, we ensure that\nour approximation achieves higher utility than traditional methods for\napplications that have asymmetric utility functions. Furthermore, in using\ndropout inference, we highlight that our new objective is identical to that of\nstandard dropout neural networks, with an additional utility-dependent penalty\nterm. We demonstrate our new loss-calibrated model with an illustrative medical\nexample and a restricted model capacity experiment, and highlight failure modes\nof the comparable weighted cross entropy approach. Lastly, we demonstrate the\nscalability of our method to real world applications with per-pixel semantic\nsegmentation on an autonomous driving data set.","url_abs":"http://arxiv.org/abs/1805.03901v1","url_pdf":"http://arxiv.org/pdf/1805.03901v1.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":"loss-calibrated-approximate-inference-in","repo_url":"https://github.com/IntelLabs/AVUC","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"autonomous-driving","task_name":"Autonomous Driving"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[{"method_slug":"dropout","method_name":"Dropout"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1805.03901","atlas_url":"https://app.syntology.ai/?focus=1805.03901","mcp":null,"developers":"https://syntology.ai/developers","held":true,"reason":"arxiv_id_in_atlas","checked_against":{"atlas_ids":549449,"atlas_ids_sha256":"afd72399dafd9906e9062927f70655952385a3d75488558e1f0484e950ab09b7","read_at":"2026-10-05T01:46:21Z"},"agent_calls":[{"tool":"get_citation_path","arguments":{"paper_1":"1805.03901"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"1805.03901\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"1805.03901"},"call":"get_concepts_for_paper(arxiv_id=\"1805.03901\")"}],"by_repo":[{"repo":"IntelLabs/AVUC","kind":"listed","samples":0,"ran":0,"instrument":0,"unverified":0}],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"ce19c870444c971aca38ec319182d277707b4707f70f9a257d95849d72421495","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}