{"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/arxiv-2605-24570","title":"PILOT: Policy-Informed Learned Optimization for Adaptive Deep Network Training","arxiv_id":"2605.24570","date":"2026-05-23","proceeding":null,"authors":["Sattam Altuuaim","Lama Ayash","Muhammad Mubashar","Naeemullah Khan"],"abstract":"Despite the central role of optimization in deep learning, most optimizers rely on update structures whose functional form is fixed before training begins. This static design can limit their ability to respond to changing gradient behavior across the loss landscape, where training may shift between stable, noisy, and inconsistent regimes. This study proposes PILOT (Policy-Informed Learned OpTimizer), an online optimizer that adapts its update behavior during training. Rather than using a fixed balance between momentum, normalization, and sign-based updates, PILOT uses gradient-direction agreement as a signal of local training stability. Conditioning the update rule on this agreement signal allows the optimizer to adjust its behavior when gradients become stable, noisy, or inconsistent. Experiments on FashionMNIST and CIFAR-10 show that PILOT consistently achieves the highest accuracy among the evaluated optimizers across convolutional settings. On the CNN architecture, PILOT reaches 94.13% on FashionMNIST and 81.94% on CIFAR-10. On ResNet-18, it further improves performance, reaching 95.71% on FashionMNIST and 93.42% on CIFAR-10. These results suggest that learning how to adapt the update structure during training can improve performance across both compact and deeper convolutional models while preserving a simple first-order optimization framework. The implementation of PILOT is publicly available at https://github.com/SattamAltwaim/PILOT.git","url_abs":"https://arxiv.org/abs/2605.24570","url_pdf":"https://arxiv.org/pdf/2605.24570","source":{"archive":null,"snapshot":"2025-07-28","note":"not in the Papers with Code archive (frozen at the snapshot)","row_kind":"graph","title_abstract_authors_date":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)"},"code_links":[],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2605.24570","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2605.24570"}},"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":[{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/SattamAltwaim/PILOT","reach":null}],"summary":{"ran":1,"ran_honours":2,"ran_violates":1,"ran_draft_wrong":1,"ran_fixture":1,"unverified":1},"by_repo_kind":{"found_in_text":{"samples":7,"ran":6,"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":"d53037b944a27b51","entry":"DiagnosticsTracker","repo":"SattamAltwaim/PILOT","repo_kind":"found_in_text","path":"pilot/optimizer.py","file_url":"https://github.com/SattamAltwaim/PILOT/blob/HEAD/pilot/optimizer.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d53037b944a27b51"}},{"code_sha256_prefix":"58f5f36553c01417","entry":"_build_phi_init","repo":"SattamAltwaim/PILOT","repo_kind":"found_in_text","path":"pilot/optimizer.py","file_url":"https://github.com/SattamAltwaim/PILOT/blob/HEAD/pilot/optimizer.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"58f5f36553c01417"}},{"code_sha256_prefix":"31ede5e92f97d8f8","entry":"_horner","repo":"SattamAltwaim/PILOT","repo_kind":"found_in_text","path":"pilot/optimizer.py","file_url":"https://github.com/SattamAltwaim/PILOT/blob/HEAD/pilot/optimizer.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"31ede5e92f97d8f8"}},{"code_sha256_prefix":"2e73f9db7b8300a6","entry":"_safe_log","repo":"SattamAltwaim/PILOT","repo_kind":"found_in_text","path":"pilot/optimizer.py","file_url":"https://github.com/SattamAltwaim/PILOT/blob/HEAD/pilot/optimizer.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"2e73f9db7b8300a6"}},{"code_sha256_prefix":"f1e8b3bf8f11451a","entry":"_safe_pow","repo":"SattamAltwaim/PILOT","repo_kind":"found_in_text","path":"pilot/optimizer.py","file_url":"https://github.com/SattamAltwaim/PILOT/blob/HEAD/pilot/optimizer.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"f1e8b3bf8f11451a"}},{"code_sha256_prefix":"b506d966d7bb2257","entry":"compute_meta_grads","repo":"SattamAltwaim/PILOT","repo_kind":"found_in_text","path":"pilot/optimizer.py","file_url":"https://github.com/SattamAltwaim/PILOT/blob/HEAD/pilot/optimizer.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b506d966d7bb2257"}},{"code_sha256_prefix":"3f35b8f6061e349a","entry":"PILOT","repo":"SattamAltwaim/PILOT","repo_kind":"found_in_text","path":"pilot/optimizer.py","file_url":"https://github.com/SattamAltwaim/PILOT/blob/HEAD/pilot/optimizer.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"3f35b8f6061e349a"}}]},"arxiv_metadata":{"licence":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)","fields":["title","abstract","authors","date"],"primary_category":"cs.LG","source":"arxiv_2026.jsonl"},"syntology_extracted_results":null}