FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3197405786> ?p ?o ?g. }

• W3197405786 abstract "The dominating NLP paradigm of training a strong neural predictor to perform one task on a specific dataset has led to state-of-the-art performance in a variety of applications (eg. sentiment classification, span-prediction based question answering or machine translation). However, it builds upon the assumption that the data distribution is stationary, ie. that the data is sampled from a fixed distribution both at training and test time. This way of training is inconsistent with how we as humans are able to learn from and operate within a constantly changing stream of information. Moreover, it is ill-adapted to real-world use cases where the data distribution is expected to shift over the course of a model's lifetime. The first goal of this thesis is to characterize the different forms this shift can take in the context of natural language processing, and propose benchmarks and evaluation metrics to measure its effect on current deep learning architectures. We then proceed to take steps to mitigate the effect of distributional shift on NLP models. To this end, we develop methods based on parametric reformulations of the distributionally robust optimization framework. Empirically, we demonstrate that these approaches yield more robust models as demonstrated on a selection of realistic problems. In the third and final part of this thesis, we explore ways of efficiently adapting existing models to new domains or tasks. Our contribution to this topic takes inspiration from information geometry to derive a new gradient update rule which alleviate catastrophic forgetting issues during adaptation." @default.
• W3197405786 created "2021-09-13" @default.
• W3197405786 creator A5001170954 @default.
• W3197405786 date "2021-09-03" @default.
• W3197405786 modified "2023-09-27" @default.
• W3197405786 title "Learning Neural Models for Natural Language Processing in the Face of Distributional Shift" @default.
• W3197405786 cites W113837456 @default.
• W3197405786 cites W131533222 @default.
• W3197405786 cites W1484551447 @default.
• W3197405786 cites W149482658 @default.
• W3197405786 cites W1552847225 @default.
• W3197405786 cites W1614298861 @default.
• W3197405786 cites W1647779468 @default.
• W3197405786 cites W1682403713 @default.
• W3197405786 cites W1753482797 @default.
• W3197405786 cites W179875071 @default.
• W3197405786 cites W1834627138 @default.
• W3197405786 cites W1862810382 @default.
• W3197405786 cites W1902237438 @default.
• W3197405786 cites W1905522558 @default.
• W3197405786 cites W1965555277 @default.
• W3197405786 cites W1968355947 @default.
• W3197405786 cites W2011588348 @default.
• W3197405786 cites W2022775778 @default.
• W3197405786 cites W2026653933 @default.
• W3197405786 cites W2033178790 @default.
• W3197405786 cites W2034368206 @default.
• W3197405786 cites W2036963181 @default.
• W3197405786 cites W2040555227 @default.
• W3197405786 cites W2047449506 @default.
• W3197405786 cites W2061873838 @default.
• W3197405786 cites W2064675550 @default.
• W3197405786 cites W2079067446 @default.
• W3197405786 cites W2092123844 @default.
• W3197405786 cites W2095705004 @default.
• W3197405786 cites W2096384330 @default.
• W3197405786 cites W2099111195 @default.
• W3197405786 cites W2099471712 @default.
• W3197405786 cites W2101105183 @default.
• W3197405786 cites W2101761627 @default.
• W3197405786 cites W2102411528 @default.
• W3197405786 cites W2104094955 @default.
• W3197405786 cites W2108598243 @default.
• W3197405786 cites W2108682071 @default.
• W3197405786 cites W2109100253 @default.
• W3197405786 cites W2111355007 @default.
• W3197405786 cites W2117278770 @default.
• W3197405786 cites W2120587290 @default.
• W3197405786 cites W2124807415 @default.
• W3197405786 cites W2127863960 @default.
• W3197405786 cites W2130158090 @default.
• W3197405786 cites W2130942839 @default.
• W3197405786 cites W2131953535 @default.
• W3197405786 cites W2133459682 @default.
• W3197405786 cites W2141440284 @default.
• W3197405786 cites W2153848201 @default.
• W3197405786 cites W2153929442 @default.
• W3197405786 cites W2162651021 @default.
• W3197405786 cites W2167729035 @default.
• W3197405786 cites W2169200297 @default.
• W3197405786 cites W2170612786 @default.
• W3197405786 cites W2183341477 @default.
• W3197405786 cites W2193413348 @default.
• W3197405786 cites W2194321275 @default.
• W3197405786 cites W2194775991 @default.
• W3197405786 cites W2243397390 @default.
• W3197405786 cites W2250342921 @default.
• W3197405786 cites W2250473257 @default.
• W3197405786 cites W2250559305 @default.
• W3197405786 cites W2251939518 @default.
• W3197405786 cites W2251994258 @default.
• W3197405786 cites W2255883267 @default.
• W3197405786 cites W2267126114 @default.
• W3197405786 cites W22861983 @default.
• W3197405786 cites W2293111166 @default.
• W3197405786 cites W2296283641 @default.
• W3197405786 cites W2336156739 @default.
• W3197405786 cites W2342840547 @default.
• W3197405786 cites W2396767181 @default.
• W3197405786 cites W2402268235 @default.
• W3197405786 cites W2432142698 @default.
• W3197405786 cites W2464152434 @default.
• W3197405786 cites W2483390977 @default.
• W3197405786 cites W2488917374 @default.
• W3197405786 cites W2511950321 @default.
• W3197405786 cites W2525778437 @default.
• W3197405786 cites W2531499355 @default.
• W3197405786 cites W2558813012 @default.
• W3197405786 cites W2560647685 @default.
• W3197405786 cites W2561274697 @default.
• W3197405786 cites W2562747313 @default.
• W3197405786 cites W2567639685 @default.
• W3197405786 cites W2577255746 @default.
• W3197405786 cites W2579599705 @default.
• W3197405786 cites W2594590228 @default.
• W3197405786 cites W2594877703 @default.
• W3197405786 cites W2595653137 @default.
• W3197405786 cites W2622263826 @default.
• W3197405786 cites W2735135478 @default.
• W3197405786 cites W2737492962 @default.

• next