[ https://issues.apache.org/jira/browse/SPARK-1585?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Patrick Wendell updated SPARK-1585: ----------------------------------- Fix Version/s: (was: 1.0.0) 1.1.0 > Not robust Lasso causes Infinity on weights and losses > ------------------------------------------------------ > > Key: SPARK-1585 > URL: https://issues.apache.org/jira/browse/SPARK-1585 > Project: Spark > Issue Type: Bug > Components: MLlib > Affects Versions: 0.9.1 > Reporter: Xusen Yin > Assignee: Xusen Yin > Fix For: 1.1.0 > > > Lasso uses LeastSquaresGradient and L1Updater, but > diff = brzWeights.dot(brzData) - label > in LeastSquaresGradient would cause too big diff, then will affect the > L1Updater, which increases weights exponentially. Small shrinkage value > cannot lasso weights back to zero then. Finally, the weights and losses reach > Infinity. > For example, data = (0.5 repeats 10k times), weights = (0.6 repeats 10k > times), then data.dot(weights) approximates 300+, the diff will be 300. Then > L1Updater sets weights to approximate 300. In the next iteration, the weights > will be set to approximate 30000, and so on. -- This message was sent by Atlassian JIRA (v6.2#6252)