Rekod bukti kaedah
Knowledge Distillation
Knowledge Distillation is a model-compression technique, introduced by Geoffrey Hinton and colleagues in 2015, that trains a small student model using the soft-label outputs of a large teacher model. Distilled models such as DistilBERT and TinyBERT reach roughly 97% of the larger model's performance while running far faster.
Rekod sumber
Petikan disalin secara verbatim daripada rekod sumber kaedah. Tiada pengesahan peringkat tuntutan disimpulkan daripadanya.
Knowledge Distillation (Teacher–Student Model Compression)
Rekod kaedah taksonomik · ml-model / deep-learning
- Hinton, G., Vinyals, O. & Dean, J. (2015). Distilling the Knowledge in a Neural Network. NeurIPS Deep Learning Workshop. · URL
- Sanh, V., Debut, L., Chaumond, J. & Wolf, T. (2019). DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter. arXiv:1910.01108. · URL
Tuntutan yang dikurasi
Tuntutan disimpan dalam lejar bukti, setiap satu dengan penilaiannya sendiri.
Tiada tuntutan terkurasi lagi
Pandangan ini tidak mencipta penilaian tuntutan apabila lejar tiada.
Kaedah berkaitan
Dijana daripada graf kaedah dan ditunjukkan sebagai perhubungan yang dicadangkan mesin — tiada tuntutan bukti disimpulkan.