Ci bi mu melGis bi ci topp
Fonction yuy ñàkk
Fondamentaal yi
GUIDE bu am solo
Liggéeyu tàmbali
Fonction activation mooy buntu yu ndaw yi nekkul ci biir neuron bu nekk, loolu mooy tax reso neuronal yi jàng motif yu xawa jafee xam, yu xawa sëgg, duñu nekk ligne yu jub kese.
Déglul tegtal biiIA-nettali · ~2 min
Ci xët wii2 simili jàng
Résumé
Without them, a deep network would collapse into a single linear equation.
Plongeur bu xóot
Neuron bu nekk dafay xayma limu dugal ci limu pondéree, waaye limu dugal ci ligneer la. Defar ay couche ligneaire yu bari, ci wàllu math, benn fonction ligneaire bu rëy kese ngay am, ak limu mëna doon. Fonction yiy aktive dañuy dindi lii ci def coppite bu amul ligne ci neuron bu nekk, loolu mooy tax reso yi mëna xayma bépp fonction. Li gëna siiw mooy ReLU, mooy génne dugal bi sudee positif la ak nul luko moy; Dafa gaaw te moytu yenn jafe-jafe tàggat ci fonction yu yàgg yi. Sigmoid ak tanh squash valeur ci rang yu yam te bariwoon nañu ci taarix waaye mën nañu am jafe-jafe gradient yuy réer ci reso yu xóot yi. Fonction softmax bi ñuy jëfandikoo ci génne gi, dafay soppi poñ yu bees yi ci séddale probabilite ci kaw klaas yi.
Doxal ci àdduna dëgg
Jëfandikoo ReLU ci biir ay layer yu nëbbu ci reso convolutionnel suko defee mu mëna jàng ay dogal yu xawa sëgg ngir xàmmee nataal
Jëfandikoo softmax ci couche bu mujj bi ngir soppi poñ yu ñor yi ci classifier bi ci probabilite classe yuy boole benn
Tann GELU yiy tàmbali liggéey ci biir xeetu làkk transformatër ngir gëna yomba def gradient
Weccoo ci Leaky ReLU sudee neuron yu bari ci reso bi dee nañu ba noppi bàyyi tontu
Ëlëgu fonction yiy tàmbali
ReLU ak cousin-am bu nooy bi GELU ñoo ëpp doole tay, GELU moo gëna am solo ci transformateur yi ndax courbe bu nooy bi dafay méngoo bu baax ak seen dynamique de formation. Gëstu dafay saytu li ñu jàng ak liñ tëj lu melni SwiGLU, leegi ñu bari ci xeeti làkk yu mag, ñuy jëfandikoo gating buy yokk ngir gëna fësal kàddu. Tendens bu yaatu bi mooy fonction yu nooy, yu am buntu yuy gëna baaxal debit gradient bi ak kalite model bi ci eskaal bi. Bi activations exotic feeñ saa yu nekk ci këyit, fonction yu yomb, yu baax dañuy gëna am ndam ci jëf ndax dañuy tàggat ci anam wu wóor ci model yu mag.
Teg ko ci jëf
Lu tax mu am solo
- dogal yu gëna leer. Daf lay jàppale nga tàqale kàddu yu leer ci wàllu xarala ak làkku fësal njaay.
- Njëgg ak budget. Mën nga laaj laaj yu gëna baax ci samp gi balaa ngay dugal xaalis wala sa jotu liggéey.
- Ekip ak def liggéey. Ekip yi bokk xam-xam ñoo gëna mëna jël yenn dogal ci wàllu produit, politik ak jàng.
Moytul
- Ekip yu bari mën nañu jëfandikoo benn baat ci anam wu wuute, kon teela leeral yaatuwaayam.
- Benchmark yi mën nañu nuru lu am doole waaye performance yi ci àdduna bi duñu tolloo.
- Bëgg kalite done ak palaŋu jàngat dafay faral di jur njariñ yu yomba dagg.
Jéego yi ci topp
- Tàmbaleel ci joxe leeral ci làkk wu leer ci njariñ li nga soxla.
- Tannal benn metric bu baax ak benn anam bu baaxul balaa ngay saytu.
- Doxal ab pilote bu ndaw ak ay done yu representatif, du ab demo bu leer.
- Bindal fi Fonction Activation di jàppale ak fi pexe yu gëna yomba gëna baax.
Weyal di banneexu
Bulletin bu amul fayda
Jàngal xibaar AI bu bes
Ñatti jaar-jaari IA yuñ firndeel bis bu nekk ci ayu-bis bi, ñu bind leen ci Àngle bu leer. Gratis ba fàww, amul yëgle.
Benn email bis bu nekk ci ayu-bis bi. Benn klik nga dindi abonemaa bi. Dunu musa jaay wala séddoo sa adres.
Nattal sa bopp
Take the Activation Functions quiz
Feedback ci saasi ci bépp tontu, ak sertifikaa buñ mëna séddoo ak ID buñ mëna saytu soo jàllee ci njàng mi.
Jàppale njàngum IA te doo fay. AI Understanding mbootaay la buy def te Yàlla tax — amul yëgle, amul paywall, musul am. Defal ab sarax
Laaj yi ñuy faral di laaj
What is Activation Functions?
Fonction activation mooy buntu yu ndaw yi nekkul ci biir neuron bu nekk, loolu mooy tax reso neuronal yi jàng motif yu xawa jafee xam, yu xawa sëgg, duñu nekk ligne yu jub kese. Suñu leen amul, reso bu xóot bi dina daanu nekk benn equation lineaire.
Why can stacking only linear layers not learn a curved decision boundary?
Without nonlinear activations, multiple linear layers compose to another linear mapping.
Why can sigmoid and tanh stall learning in deep networks when inputs reach their extremes?
At their flat extremes, their near-zero gradients weaken the backpropagated signal.
A multiclass classifier needs output values that sum to one across classes. Which function does the guide use?
Softmax turns raw class scores into a probability distribution over classes.
What does the “dying ReLU” problem mean in the guide?
A ReLU stuck receiving negative inputs can keep outputting zero, preventing useful updates through that unit.
Which variant is suggested when many ReLU units stay inactive?
The guide names Leaky ReLU as a response to dead units because its negative side is not flat at zero.
Weyal di jàng
Gid yu jëm ci loolu
Tann nañu yeneen njiit ngir topic bii