Unbiased Online Recurrent Optimization

@article{Tallec2018UnbiasedOR,
  title={Unbiased Online Recurrent Optimization},
  author={Corentin Tallec and Yann Ollivier},
  journal={CoRR},
  year={2018},
  volume={abs/1702.05043}
}
The novel Unbiased Online Recurrent Optimization (UORO) algorithm allows for online learning of general recurrent computational graphs such as recurrent network models. It works in a streaming fashion and avoids backtracking through past activations and inputs. UORO is computationally as costly as Truncated Backpropagation Through Time (truncated BPTT), a widespread algorithm for online learning of recurrent networks Jaeger (2002). UORO is a modification of NoBackTrack Ollivier et al. (2015… CONTINUE READING
Tweets
This paper has been referenced on Twitter 37 times. VIEW TWEETS

Citations

Publications citing this paper.
SHOWING 1-10 OF 19 CITATIONS

References

Publications referenced by this paper.
SHOWING 1-10 OF 18 REFERENCES

Tutorial on training recurrent neural networks, covering BPPT, RTRL, EKF and the “echo state network

  • Herbert Jaeger
  • 2002
Highly Influential
10 Excerpts

Similar Papers

Loading similar papers…