This paper studies how joint training of multiple support vector machines (SVMs) can improve the effectiveness and efficiency of automatic image annotation. We cast image annotation as an output-related multi-task learning framework, with the prediction of each tag’s presence as one individual task. Evidently, these tasks are related via dependencies between tags. The proposed joint learning fr...

the purpose of this study was to investigate the impact of weblog task assessment on iranian efl learners’ reading ability. to this purpose, a group of 30 english language students (both male and female) in rozaneh language institute in najafabad were selected based on their performance on the opt. the researchers tried to select a homogeneous sample in terms of age factor and proficiency level...

This paper studies uncertainty using the ECB Survey of Professional Forecasters’ data. We consider both inflation and real GDP growth forecasts at the micro level and explore forecast uncertainty using two alternative measures, i.e. conventional standard deviation of individual point forecasts and the median values of individual forecasters’ uncertainty, which are based on subjective probabilit...

We present preliminary results on extending Model-Agnostic Meta-Learning (MAML) (Finn et al., 2017a) to fast adaptation to new classification tasks in the presence of unlabeled data. Using synthetic data, we show that MAML can adapt to new tasks without any labeled examples (unsupervised adaptation) when the new task has the same output space (classes) as the training tasks do. We further exten...

The specific aims of this study were to: 1) investigate the encoding of forelimb muscle activity timing and magnitude by corticomotoneuronal (CM) cells, 2) test the stability of primary motor cortex (M1) output to forelimb muscles under different task conditions, and 3) characterize input/output relationships associated with different intracortical microstimulation (ICMS) methods. Neuronal reco...