Sekvensmodeller og generative modeller

103 metoder i denne familie.

Udvalgte

OpmærksomhedsmekanismeThe attention mechanism, introduced by Bahdanau, Cho and Bengio in 2015 and refined by Luong, Pham and Manning the same year, lets a sequence decoder dynamically learn which of the AutoencoderAn autoencoder is an encoder-decoder neural network, popularised by Hinton and Salakhutdinov in 2006, that compresses data into a low-dimensional latent code and then reconstructs Bidirectional RNNA Bidirectional RNN, introduced by Schuster and Paliwal in 1997, processes a sequence in both forward and backward directions so that every position has access to its full surround CrossformerCrossformer is a Transformer-based architecture for multivariate time series forecasting, introduced by Yunhao Zhang and Junchi Yan at ICLR 2023. Unlike earlier Transformer variant CycleGAN: Uparret billed-til-billed-oversættelse med cyklisk konsistensCycleGAN, introduced by Zhu et al. at ICCV 2017, learns to translate images between two visual domains without requiring paired training examples. It trains two generators and two DeepARDeepAR is Amazon's industrial forecasting model, introduced by Salinas, Flunkert and Gasthaus (2017; published 2020), that uses an autoregressive recurrent neural network to estima

Læsesti

Dette emnes mest refererede grundlæggende metoder, i den rækkefølge de blev udviklet — et godt sted at begynde, hvis du er ny her.

Recurrent Neural Network1986–1990af Rumelhart, D. E.; Elman, J. L.
Long Short-Term Memory (LSTM)1997af Hochreiter, S. & Schmidhuber, J.
Gated Recurrent Unit (GRU)2014af Cho, K., van Merrienboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., & Bengio, Y.
Generativ modstridende netværk2014af Goodfellow, I. et al.
Variational Autoencoder2014af Kingma, D. P. & Welling, M.
Fintunet Transformer2017–2019af Vaswani et al. (architecture); fine-tuning paradigm popularised by Howard & Ruder, Devlin et al.
Multimodal Transformer2019–2021af Lu et al. (ViLBERT); Radford et al. (CLIP)
Vision Transformer2021af Dosovitskiy, A. et al.

alle metoder på denne hylde ↓

Alle metoder 103

Opmærksomhedsmekanisme Autoencoder Bidirectional RNN Crossformer CycleGAN: Uparret billed-til-billed-oversættelse med cyklisk konsistens DeepAR Diffusionsmodel Domæne-adaptiv diffusionsmodel Domæneadaptiv GAN Domæne-adaptiv GRU Domæneadaptivt Rekurrent Neuralt Netværk Domæne-adaptive sætningsindlejringer Domæne-adaptiv Transformer Domæneadaptiv Variational Autoencoder Domæne-adaptiv Vision Transformer Forklarlig Diffusionsmodel Forklarlig GAN Forklarlig GRU Forklarlig LSTM Forklarlig Rekurrent Neuralt Netværk Forklarlig Transformer Forklarlig Variational Autoencoder FEDformer: Frekvensforstærket Dekomponeret Transformer Finjusteret diffusionsmodel Finetunet Generativt Adversarialt Netværk Finjusteret GRU Fintunet LSTM Finetunet Recurrent Neural Network Finetunet tekstresumé Fintunet Transformer Finjusteret Variational Autoencoder Finetunet Vision Transformer Gated Recurrent Unit (GRU)Generativ modstridende netværk Graph Attention Network Gated Recurrent Unit (GRU)Informer iTransformer: Inverteret Transformer til Multivariat Tidsserieprognose Latente diffusionsmodeller Long Short-Term Memory (LSTM)Longformer / BigBird LSTM Maskerede Autoencoders Moirai: Universal Transformer til Tidsserieprognoser Multilingual Diffusionsmodel Multilingual GAN Multilingual GRU Multilingual LSTM Multilingual Recurrent Neural Network Flersproget Tekstopsummering Multilingual Variational Autoencoder Multilingual Vision Transformer Multimodal Diffusionsmodel Multimodal GAN Multimodal GRU Multimodal LSTM Multimodal Recurrent Neural Network Multimodal Transformer Multimodal Variational Autoencoder Multimodal Vision Transformer Non-stationary Transformer PatchTST Pyraformer: Pyramidal Attention Transformer til langtids-tidsserie-prognoser Recurrent Neural Network Reformer: Den effektive Transformer til lange sekvenser Score-baseret generativ model SegRNN: Segment Recurrent Neural Network til Langtids Tidsserieprognoser Multi-Head Self-Attention Selv-superviseret diffusionsmodel Selvovervåget GAN Selv-overvåget GRU Selv-overvåget Transformer Selv-overvåget Variational Autoencoder Selv-overvåget Vision Transformer Semiovervåget diffusionsmodel Semi-supervised GAN Semi-superviseret GRU Semi-supervised LSTM Semi-superviseret Transformer Semi-supervised Variational Autoencoder Semi-supervised Vision Transformer Sekvens-til-sekvens-model Swin Transformer T5 (Text-to-Text Transfer Transformer)Temporal Fusion Transformer Time-MoE: Mixture-of-Experts Tidsrække Fundamentmodel TiRex: Zero-Shot Tidsrække-prognoser med xLSTM Transfer Learning GAN Overførselslæring med Variational Autoencoder Transfer Learning med Diffusionsmodel Transferindlæring med LSTM Overførselslæring med Recurrent Neural Network Variational Autoencoder Vision Transformer Wasserstein GAN (WGAN)Svagt superviseret diffusionsmodel Svagt superviseret GAN Svagt overvåget GRU Svagt superviseret LSTM Svagt superviseret recurrent neuralt netværk Svagt superviseret Transformer Svagt overvåget Variational Autoencoder Svagt superviseret Vision Transformer

Mere i Dyb læring

Arkitekturer og træning 124