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 > Isaac Newton Institute Seminar Series > On computational barriers in data science and the paradoxes of deep learning

## On computational barriers in data science and the paradoxes of deep learningAdd to your list(s) Download to your calendar using vCal - Anders Hansen (University of Cambridge)
- Thursday 02 November 2017, 11:10-12:00
- Seminar Room 1, Newton Institute.
If you have a question about this talk, please contact INI IT. VMVW02 - Generative models, parameter learning and sparsity The use of regularisation techniques such as l^1 and Total Variation in Basis Pursuit and Lasso, as well as linear and semidefinite programming and neural networks (deep learning) has seen great success in data science. Yet, we will discuss the following paradox: it is impossible to design algorithms to find minimisers accurately for these problems when given inaccurate input data, even when the inaccuracies can be made arbitrarily small. The paradox implies that any algorithm designed to solve these problems will fail in the following way: For fixed dimensions and any small accuracy parameter epsilon > 0, one can choose an arbitrary large time T and find an input such that the algorithm will run for longer than T and still not have reached epsilon accuracy. Moreover, it is impossible to determine when the algorithm should halt to achieve an epsilon accurate solution. The largest epsilon for which this failure happens is called the Breakdown-epsilon. Typically, the Breakdown-epsilon > 1/2 even when the the input is bounded by one, is well-conditioned, and the objective function can be computed with arbitrary accuracy. This talk is part of the Isaac Newton Institute Seminar Series series. ## This talk is included in these lists:- All CMS events
- Featured lists
- INI info aggregator
- Isaac Newton Institute Seminar Series
- School of Physical Sciences
- Seminar Room 1, Newton Institute
- bld31
Note that ex-directory lists are not shown. |
## Other listsSt Edmund's College Political Forum SECPF Scott Polar Research Institute - HCEP (Histories, Cultures, Environments and Politics) Research Seminars Engineering Safe AI## Other talksAlgorithmic Investigation of Large Biological Data sets Chains and Invisible Threads: Marx on Republican Liberty and Domination Holonomic D-modules, b-functions, and coadmissibility Overview of Research Process Bayesian optimal design for Gaussian process model The homelands of the plague: Soviet disease ecology in Central Asia, 1920s–1950s |