We analyze classification error on unseen cases, i.e. cases that are different from those in the training set. Unlike standard generalization error, this off-training-set error may differ significantly from the empirical error with high probability even with large sample sizes. We derive a datadependent bound on the difference between off-training-set and standard generalization error. Our resu...

Forward error correction codes (FEC) are used for error detection and correction in communication systems. Low density parity check code (LDPC) is used as a powerful Forward Error Correction code in long distance communication systems which works close to the Shannon limit. Unlike other conventional channel code, the decoding algorithm used for LDPC codes is an iterative message passing algorit...

We address the problem of detecting English language learner errors by using a discriminative high-order sequence model. Unlike most work in error-detection, this method is agnostic as to specific error types, thus potentially allowing for higher recall across different error types. The approach integrates features from many sources into the error-detection model, ranging from language model-ba...

Validation studies that compare survey-reported earnings to administrative-recorded earnings are useful to assess the extent and implications of measurement error in labor market data. While previous work typically used small restrictive samples with topcoded earnings, this paper uses data from the 1996 Survey of Income and Program Participation (SIPP) Panel matched to Social Security administr...

Using a perturbation theory with a hard-sphere reference system we have directly calculated free energies of fluid and solid phases of aluminum with an embedded atom model potential. Unlike other approaches such as thermodynamic integration, we do not require any simulations. Moreover, the free energies of the two different phases are calculated in a single approach, unlike approximations like ...