{"id":"W4399715009","doi":"10.1007/978-3-031-62700-2_14","title":"Enhancing Temporal Transformers for Financial Time Series via Local Surrogate Interpretability","year":2024,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Stock Market Forecasting Methods","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Ottawa","funders":"","keywords":"Interpretability; Computer science; Transformer; Series (stratigraphy); Finance; Artificial intelligence; Electrical engineering; Business; Engineering","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.01309418,0.0007029107,0.001111219,0.001158206,0.0003391676,0.0007081971,0.002227074,0.000527591,0.0002760137],"category_scores_gemma":[0.004003905,0.0005503646,0.0005270614,0.001077639,0.002240968,0.0006526485,0.0007013809,0.001032463,0.0001519657],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005036236,"about_ca_system_score_gemma":0.00110283,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003079057,"about_ca_topic_score_gemma":0.0007765438,"domain_scores_codex":[0.9930372,0.0001360561,0.00147713,0.002423626,0.002045976,0.0008800699],"domain_scores_gemma":[0.9914666,0.006433267,0.0003695868,0.0008955974,0.0005986333,0.0002362532],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.000124441,0.00001166815,0.00005072578,0.00007793833,0.00001037078,0.00002544818,0.0008875822,0.002227493,0.0002612464,0.0005994376,0.00004580073,0.9956778],"study_design_scores_gemma":[0.0002267948,0.0004615552,0.0000718115,0.0005790934,0.00002903013,0.00008829111,0.000001341025,0.2603748,0.005576927,0.7235872,0.008232539,0.0007705981],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0003990876,0.0002504173,0.9893089,0.0006396792,0.004675047,0.0008621131,0.00006304171,0.0001412314,0.003660517],"genre_scores_gemma":[0.3383589,0.000007224918,0.6498555,0.0007743465,0.001442623,0.00006467772,0.00001706433,0.0001437592,0.009335934],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9949073,"threshold_uncertainty_score":0.9996948,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03428577947328498,"score_gpt":0.3330686163089388,"score_spread":0.2987828368356538,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}