Matches in SemOpenAlex for { <https://semopenalex.org/work/W3195630663> ?p ?o ?g. }
- W3195630663 abstract "Tensor networks are efficient representations of high-dimensional tensors with widespread applications in quantum many-body physics. Recently, they have been adapted to the field of machine learning, giving rise to an emergent research frontier that has attracted considerable attention. Here, we study the trainability of tensor-network based machine learning models by exploring the landscapes of different loss functions, with a focus on the matrix product states (also called tensor trains) architecture. In particular, we rigorously prove that barren plateaus (i.e., exponentially vanishing gradients) prevail in the training process of the machine learning algorithms with global loss functions. Whereas, for local loss functions the gradients with respect to variational parameters near the local observables do not vanish as the system size increases. Therefore, the barren plateaus are absent in this case and the corresponding models could be efficiently trainable. Our results reveal a crucial aspect of tensor-network based machine learning in a rigorous fashion, which provide a valuable guide for both practical applications and theoretical studies in the future." @default.
- W3195630663 created "2021-08-30" @default.
- W3195630663 creator A5034703407 @default.
- W3195630663 creator A5037109770 @default.
- W3195630663 creator A5057614578 @default.
- W3195630663 creator A5061987441 @default.
- W3195630663 date "2022-12-28" @default.
- W3195630663 modified "2023-10-07" @default.
- W3195630663 title "Presence and Absence of Barren Plateaus in Tensor-Network Based Machine Learning" @default.
- W3195630663 cites W1492999010 @default.
- W3195630663 cites W1653517664 @default.
- W3195630663 cites W1840046823 @default.
- W3195630663 cites W1848366941 @default.
- W3195630663 cites W1922629554 @default.
- W3195630663 cites W1965555277 @default.
- W3195630663 cites W1988369744 @default.
- W3195630663 cites W2003333523 @default.
- W3195630663 cites W2015775929 @default.
- W3195630663 cites W2021115151 @default.
- W3195630663 cites W2035298342 @default.
- W3195630663 cites W2044090207 @default.
- W3195630663 cites W2044891881 @default.
- W3195630663 cites W2070375169 @default.
- W3195630663 cites W2071389404 @default.
- W3195630663 cites W2092144622 @default.
- W3195630663 cites W2094366212 @default.
- W3195630663 cites W2116102644 @default.
- W3195630663 cites W2139778726 @default.
- W3195630663 cites W2217912240 @default.
- W3195630663 cites W2252795400 @default.
- W3195630663 cites W2337082154 @default.
- W3195630663 cites W2414456771 @default.
- W3195630663 cites W2419175238 @default.
- W3195630663 cites W2516041031 @default.
- W3195630663 cites W2521267242 @default.
- W3195630663 cites W2559394418 @default.
- W3195630663 cites W2582157661 @default.
- W3195630663 cites W2582761306 @default.
- W3195630663 cites W2583335063 @default.
- W3195630663 cites W2606578469 @default.
- W3195630663 cites W2607839392 @default.
- W3195630663 cites W2617994470 @default.
- W3195630663 cites W2618945100 @default.
- W3195630663 cites W2750586355 @default.
- W3195630663 cites W2750673150 @default.
- W3195630663 cites W2753545915 @default.
- W3195630663 cites W2762508157 @default.
- W3195630663 cites W2792315573 @default.
- W3195630663 cites W2792946961 @default.
- W3195630663 cites W2794444783 @default.
- W3195630663 cites W2794727009 @default.
- W3195630663 cites W2798945316 @default.
- W3195630663 cites W2903891684 @default.
- W3195630663 cites W2912516940 @default.
- W3195630663 cites W2923537029 @default.
- W3195630663 cites W2942828225 @default.
- W3195630663 cites W2949570027 @default.
- W3195630663 cites W2949757588 @default.
- W3195630663 cites W2971476734 @default.
- W3195630663 cites W2979788593 @default.
- W3195630663 cites W3000624483 @default.
- W3195630663 cites W3000699591 @default.
- W3195630663 cites W3015649859 @default.
- W3195630663 cites W3015983231 @default.
- W3195630663 cites W3030829226 @default.
- W3195630663 cites W3033945399 @default.
- W3195630663 cites W3087264907 @default.
- W3195630663 cites W3090921460 @default.
- W3195630663 cites W3091811465 @default.
- W3195630663 cites W3095463428 @default.
- W3195630663 cites W3097990818 @default.
- W3195630663 cites W3098768946 @default.
- W3195630663 cites W3099910066 @default.
- W3195630663 cites W3099956647 @default.
- W3195630663 cites W3100068159 @default.
- W3195630663 cites W3100993774 @default.
- W3195630663 cites W3102320711 @default.
- W3195630663 cites W3102352811 @default.
- W3195630663 cites W3103713775 @default.
- W3195630663 cites W3103870741 @default.
- W3195630663 cites W3103872322 @default.
- W3195630663 cites W3103945605 @default.
- W3195630663 cites W3104022488 @default.
- W3195630663 cites W3104481216 @default.
- W3195630663 cites W3105432754 @default.
- W3195630663 cites W3105870134 @default.
- W3195630663 cites W3112028493 @default.
- W3195630663 cites W3119774682 @default.
- W3195630663 cites W3125452792 @default.
- W3195630663 cites W3134405558 @default.
- W3195630663 cites W3136233239 @default.
- W3195630663 cites W3149637861 @default.
- W3195630663 cites W3156188855 @default.
- W3195630663 cites W3163320491 @default.
- W3195630663 cites W3165717557 @default.
- W3195630663 cites W3169377977 @default.
- W3195630663 cites W3182433019 @default.
- W3195630663 cites W3183968998 @default.
- W3195630663 cites W3198049459 @default.
- W3195630663 cites W3204404766 @default.