{"id":"6d1e75ef-74c3-4126-98c2-7fa69a44bba0","arxiv_id":"1908.00399","paper_version":4,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"A kernelized inverse-optimization approach forecasts EV-fleet charging and discharging power and derives market bid curves, outperforming support vector and ridge regression on synthetic case studies.","lead":"This paper builds a two-step machine-learning method that forecasts how a fleet of electric vehicles will charge or discharge in response to electricity prices, and turns that forecast into a bid curve for the power market. It is a practical idea for aggregators who need both a prediction and a market offer, though the tests use simulated fleet data rather than real operations.","discovery_kind":"new_method","skeptic_critique":null,"referee_report":null,"author_rebuttal":null,"desk_editor":null,"rs_alignment":null,"lean_confirmation":null,"pith_extraction":null,"created_at":"2026-08-14T15:59:14.351959+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":null,"supporting_citations":[],"review_version":1}