Decoding is deciding under uncertainty — the case of NMT

Add to your list(s) Download to your calendar using vCal

• Bryan Eikema (University of Amsterdam)
• Friday 11 November 2022, 12:00-13:00
• Virtual (Zoom).

If you have a question about this talk, please contact Michael Schlichtkrull.

Abstract:

In neural machine translation (NMT), we search for the mode of the model distribution to form predictions. We do so mostly following the intuition that the most probable outcome ought to be an important summary of the distribution. Despite our intuition, there’s plenty of evidence against the adequacy of the most probable translations in NMT . In this talk, I make a case to move away from mode-seeking search as a tool for decision making as well as for model criticism. I will highlight reasons concerning MT as a task, NMT as a probabilistic model, and MLE as training algorithm. Finally, I’ll turn to statistical decision theory and motivate a different rule for making decisions, one which is familiar to statistical MT folks like those of my generation and earlier, as well as a modern approximation of it. I’ll close the talk with a discussion of merits and limitations of this decision rule, and comments on opportunities moving forward with or without mode-seeking search.

Bio:

Bryan Eikema is a PhD student at the Institute for Logic, Language, and Computation at the University of Amsterdam. His interests lie at the intersection of natural language processing and probabilistic modelling. In particular he works on inducing latent structure in parallel data and improving neural machine translation through better probabilistic modelling. His research is part of the European GoURMET project. His thesis adviser is dr. Wilker Ferreira Aziz.

Please note: The speaker for this talk has been changed due to covid; the content should be the same.

Topic: NLIP Seminar Time: Nov 11, 2022 12:00 PM London

Join Zoom Meeting https://cl-cam-ac-uk.zoom.us/j/91073515866?pwd=UnJmTER6dmZLeWpPOUo0VUJBOGxYQT09

Meeting ID: 910 7351 5866 Passcode: 646960

This talk is part of the NLIP Seminar Series series.

This talk is included in these lists:

• All Talks (aka the CURE list)
• Cambridge Centre for Data-Driven Discovery (C2D3)
• Cambridge Forum of Science and Humanities
• Cambridge Language Sciences
• Cambridge talks
• Chris Davis' list
• Department of Computer Science and Technology talks and seminars
• Graduate-Seminars
• Guy Emerson's list
• Interested Talks
• Language Sciences for Graduate Students
• NLIP Seminar Series
• PMRFPS's
• Simon Baker's List
• Trust & Technology Initiative - interesting events
• Virtual (Zoom)
• bld31
• ndk22's list
• ob366-ai4er
• rp587

Note that ex-directory lists are not shown.