# Meta-learners' learning dynamics are unlike learners'

@article{Rabinowitz2019MetalearnersLD, title={Meta-learners' learning dynamics are unlike learners'}, author={Neil C. Rabinowitz}, journal={ArXiv}, year={2019}, volume={abs/1905.01320} }

- Published in ArXiv 2019

Meta-learning is a tool that allows us to build sample-efficient learning systems. Here we show that, once meta-trained, LSTM Meta-Learners aren’t just faster learners than their sampleinefficient deep learning (DL) and reinforcement learning (RL) brethren, but that they actually pursue fundamentally different learning trajectories. We study their learning dynamics on three sets of structured tasks for which the corresponding learning dynamics of DL and RL systems have been previously described… CONTINUE READING

#### Citations

##### Publications citing this paper.

SHOWING 1-2 OF 2 CITATIONS

## A type of generalization error induced by initialization in deep neural networks

VIEW 1 EXCERPT

CITES BACKGROUND

#### References

##### Publications referenced by this paper.

SHOWING 1-10 OF 112 REFERENCES

## On the Spectral Bias of Deep Neural Networks

VIEW 10 EXCERPTS

HIGHLY INFLUENTIAL

## Training behavior of deep neural network in frequency domain

VIEW 10 EXCERPTS

HIGHLY INFLUENTIAL

## Exact solutions to the nonlinear dynamics of learning in deep linear neural networks

VIEW 9 EXCERPTS

HIGHLY INFLUENTIAL

## A mathematical theory of semantic development in deep neural networks

VIEW 4 EXCERPTS

HIGHLY INFLUENTIAL

## An analytic theory of generalization dynamics and transfer learning in deep linear networks

VIEW 4 EXCERPTS

HIGHLY INFLUENTIAL

## Meta-Learning by Adjusting Priors Based on Extended PAC-Bayes Theory

VIEW 4 EXCERPTS

HIGHLY INFLUENTIAL

## On the Information Bottleneck Theory of Deep Learning

VIEW 4 EXCERPTS

HIGHLY INFLUENTIAL

## Recasting Gradient-Based Meta-Learning as Hierarchical Bayes

VIEW 6 EXCERPTS

HIGHLY INFLUENTIAL

## The Physical Systems Behind Optimization Algorithms

VIEW 4 EXCERPTS

HIGHLY INFLUENTIAL

## Understanding training and generalization in deep learning by Fourier analysis

VIEW 2 EXCERPTS

HIGHLY INFLUENTIAL