TY - JOUR AU - Pau Rodriguez AU - Miguel Angel Bautista AU - Sergio Escalera AU - Jordi Gonzalez PY - 2018// TI - Beyond Oneshot Encoding: lower dimensional target embedding T2 - IMAVIS JO - Image and Vision Computing SP - 21 EP - 31 VL - 75 KW - Error correcting output codes KW - Output embeddings KW - Deep learning KW - Computer vision N2 - Target encoding plays a central role when learning Convolutional Neural Networks. In this realm, one-hot encoding is the most prevalent strategy due to its simplicity. However, this so widespread encoding schema assumes a flat label space, thus ignoring rich relationships existing among labels that can be exploited during training. In large-scale datasets, data does not span the full label space, but instead lies in a low-dimensional output manifold. Following this observation, we embed the targets into a low-dimensional space, drastically improving convergence speed while preserving accuracy. Our contribution is two fold: (i) We show that random projections of the label space are a valid tool to find such lower dimensional embeddings, boosting dramatically convergence rates at zero computational cost; and (ii) we propose a normalized eigenrepresentation of the class manifold that encodes the targets with minimal information loss, improving the accuracy of random projections encoding while enjoying the same convergence rates. Experiments on CIFAR-100, CUB200-2011, Imagenet, and MIT Places demonstrate that the proposed approach drastically improves convergence speed while reaching very competitive accuracy rates. UR - https://doi.org/10.1016/j.imavis.2018.04.004 L1 - http://refbase.cvc.uab.es/files/RBE2018.pdf UR - http://dx.doi.org/10.1016/j.imavis.2018.04.004 N1 - ISE; HuPBA; 600.098; 602.133; 602.121; 600.119 ID - Pau Rodriguez2018 ER -