• Vrushang Prakashkumar Anand
• Vaibhav V. Patel

A major goal of unsupervised learning is to get data representations that are useful for subsequent tasks, without access to supervised labels during training. Typically, this involves minimizing a surrogate objective, like the negative log likelihood of a generative model, with the hope that representations useful for subsequent tasks will arise as a side effect. during this work, we propose instead to directly target later desired tasks by meta-learning an unsupervised learning rule which results in representations useful for those tasks. Specifically, we target semi-supervised classification performance, and that an algorithm – an unsupervised weight update rule – that produces representations useful for this task. Additionally, we constrain our unsupervised update rule to a be a biologically-motivated, neuron-local function, which enables it to generalize to different neural network architectures, data-sets, and data modalities. We show that the meta-learned update rule produces useful features and sometimes outperforms existing unsupervised learning techniques. We further show that the meta-learned unsupervised update rule generalizes to coach networks with different widths, depths, and non-linearities. It also generalizes to coach on data with randomly permuted input dimensions and even generalizes from image data-sets to a text task.

Artificial Intelligence, Supervised Learning, Neural Network, Architectures, Datasets.

"A Survey of Artificial Intelligence for Unsupervised Learnig", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 6, page no.744-748, June 2019, Available :http://www.jetir.org/papers/JETIR1908715.pdf

"A Survey of Artificial Intelligence for Unsupervised Learnig", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 6, page no. pp744-748, June 2019, Available at : http://www.jetir.org/papers/JETIR1908715.pdf