Mediated Uncoupled Learning: Learning Functions without Direct Input-output Correspondences

by   Ikko Yamane, et al.

Ordinary supervised learning is useful when we have paired training data of input X and output Y. However, such paired data can be difficult to collect in practice. In this paper, we consider the task of predicting Y from X when we have no paired data of them, but we have two separate, independent datasets of X and Y each observed with some mediating variable U, that is, we have two datasets S_X = {(X_i, U_i)} and S_Y = {(U'_j, Y'_j)}. A naive approach is to predict U from X using S_X and then Y from U using S_Y, but we show that this is not statistically consistent. Moreover, predicting U can be more difficult than predicting Y in practice, e.g., when U has higher dimensionality. To circumvent the difficulty, we propose a new method that avoids predicting U but directly learns Y = f(X) by training f(X) with S_X to predict h(U) which is trained with S_Y to approximate Y. We prove statistical consistency and error bounds of our method and experimentally confirm its practical usefulness.


page 1

page 2

page 3

page 4


Deep Learning with Inaccurate Training Data for Image Restoration

In many applications of deep learning, particularly those in image resto...

Learning Paired-associate Images with An Unsupervised Deep Learning Architecture

This paper presents an unsupervised multi-modal learning system that lea...

JacNet: Learning Functions with Structured Jacobians

Neural networks are trained to learn an approximate mapping from an inpu...

Learning Bidirectional Action-Language Translation with Limited Supervision and Incongruent Input

Human infant learning happens during exploration of the environment, by ...

Cluster, Classify, Regress: A General Method For Learning Discountinous Functions

This paper presents a method for solving the supervised learning problem...

Deep-learning to Predict Perceived Differences to a Hidden Reference Image

Image metrics predict the perceived per-pixel difference between a refer...

Learning with Subset Stacking

We propose a new algorithm that learns from a set of input-output pairs....

Please sign up or login with your details

Forgot password? Click here to reset