AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Target Propagation in Recurrent Neural Networks

Nikolay Manchev, Michael Spratling

JMLR 2020 · Architectures and efficiency › RNNs and normalization · approach: Efficient recurrent architectures

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace BPTT with target propagation (TPTT)SRN with orthogonal init · Random Permutationmax sequence length: 70 -> 300Improved
replace BPTT with TPTTSRN · Temporal Ordermax sequence length: 120 -> 150Improved
replace BPTT with TPTTSRN · 3-bit Temporal Ordermax sequence length: 70 -> 150Improved
replace BPTT with TPTTSRN · Addingmax sequence length: 90 -> 60Hurt
replace BPTT with TPTT (best params)SRN · Temporal Order tmax=120iterations to convergence: 46,500 -> 1,600Improved
replace BPTT with TPTT common modelSRN · Random Permutation tmax=70iterations to convergence: 12,200 -> 12,700No clear effect
add Gaussian noise injection when learning the inverse G (ablation)TPTT · Adding tmax=70accuracy (%): 82.67 -> 94.99Improved
add Gaussian noise injection (ablation)TPTT · Adding tmax=90accuracy (%): 92.74 -> 96.52Improved
replace BPTT with TPTT, test on 2x longer sequencesSRN · Random Permutation (train 100, test 200)accuracy (%): 51.52 -> 100.00Improved
replace BPTT with TPTT, test on 2x longer sequencesSRN · Adding (train 100, test 200)accuracy (%): 65.63 -> 66.23No clear effect
replace TPTT with LSTMTPTT · Adding (train 100, test 200)accuracy (%): 66.23 -> 91.90Improved
increase hidden size Whh from 64 to 1024 (ablation)TPTT · MNIST sequence of pixelsmaximal accuracy (%): 49.49 -> 74.02Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy