COOKIES: By using this website you agree that we can place Google Analytics Cookies on your device for performance monitoring. |

University of Cambridge > Talks.cam > Machine Learning @ CUED > Nonlinear ICA using temporal structure: a principled framework for unsupervised deep learning

## Nonlinear ICA using temporal structure: a principled framework for unsupervised deep learningAdd to your list(s) Download to your calendar using vCal - Prof. Aapo Hyvarinen
- Thursday 19 October 2017, 11:00-12:00
- CBL Room BE-438, Department of Engineering.
If you have a question about this talk, please contact Dr R.E. Turner. Unsupervised learning, in particular learning general nonlinear representations, is one of the deepest problems in machine learning. Estimating latent quantities in a generative model provides a principled framework, and has been successfully used in the linear case, e.g. with independent component analysis (ICA) and sparse coding. However, extending ICA to the nonlinear case has proven to be extremely difficult: A straight-forward extension is unidentifiable, i.e. it is not possible to recover those latent components that actually generated the data. Here, we show that this problem can be solved by using temporal structure. We formulate two generative models in which the data is an arbitrary but invertible nonlinear transformation of time series (components) which are statistically independent of each other. Drawing from the theory of linear ICA , we formulate two distinct classes of temporal structure of the components which enable identification, i.e. recovery of the original independent components. We show that in both cases, the actual learning can be performed by ordinary neural network training where only the input is defined in an unconventional manner, making software implementations trivial. We can rigorously prove that after such training, the units in the last hidden layer will give the original independent components. [With Hiroshi Morioka, published at NIPS2016 and AISTATS2017 .] This talk is part of the Machine Learning @ CUED series. ## This talk is included in these lists:- Seminar
- All Talks (aka the CURE list)
- Biology
- CBL Room BE-438, Department of Engineering
- CBL important
- Cambridge Big Data
- Cambridge Forum of Science and Humanities
- Cambridge Language Sciences
- Cambridge Neuroscience Seminars
- Cambridge University Engineering Department Talks
- Centre for Smart Infrastructure & Construction
- Creating transparent intact animal organs for high-resolution 3D deep-tissue imaging
- Featured lists
- Guy Emerson's list
- Inference Group Summary
- Information Engineering Division seminar list
- Joint Machine Learning Seminars
- Life Science
- Life Sciences
- Machine Learning @ CUED
- Machine Learning Summary
- Neuroscience
- Neuroscience Seminars
- Neuroscience Seminars
- Required lists for MLG
- School of Technology
- Stem Cells & Regenerative Medicine
- dh539
- ndk22's list
- rp587
Note that ex-directory lists are not shown. |
## Other listsMongolia & Inner Asia Studies Unit Seminar Series Bio-Inspried Robotics Lab (BIRL) Seminar Series Marshall Holiday Lectures## Other talksWhat Sea Ice Physics is Missing from Models? Visual cognition How memory guides value-based decisions Rothschild Lecture: Image Compression with Differential Equations Prescribing step counts in type 2 diabetes and hypertension:Results of the Step Monitoring to improve ARTERial health trail Towards bulk extension of near-horizon geometries |