Idanwo ti o sopọ mọ itọsọna · Alabọde Ipele

Idanwo Awọn sẹẹli Iranti Igba Kukuru gigun

Test LSTM gate roles, cell-state retention, input and output control, additive updates, gradient flow, and sequence limits.

Jẹmọ itọsọna awọn ọnaAwọn sẹẹli Iranti Igba kukuru gigun
Ibeere 1 ti 8

In a common modern three-gate LSTM variant, which gates regulate cell information and exposed output?

Awotẹlẹ gbogbo ibeere8 ibeere

Awọn idahun ati awọn alaye han bi o ṣe n mu ibeere naa.

  1. In a common modern three-gate LSTM variant, which gates regulate cell information and exposed output?
    • Ka, kọ, ati ṣiṣẹ awọn ẹnu-ọna
    • Bẹrẹ, da duro, ati daduro awọn ilẹkun
    • Gbagbe, titẹ sii, ati awọn ẹnu-ọna ti o jade
    • Fi koodu sii, ṣe koodu, ati lọ si awọn ẹnu-ọna
  2. What training difficulty was LSTM designed to mitigate?
    • Awọn paramita pupọ pupọ lati fipamọ
    • Vanishing gradients during learning over long sequences
    • Ailagbara lati lo GPU kan
    • Aini iṣẹ imuṣiṣẹ
  3. Tani o ṣe agbekalẹ LSTM, ati ni ọdun wo?
    • Vaswani ati awọn ẹlẹgbẹ ni ọdun 2017
    • Rumelhart ati Hinton ni ọdun 1986
    • LeCun ati Bengio ni ọdun 2012
    • Hochreiter ati Schmidhuber ni ọdun 1997
  4. In a common LSTM cell update, why can the additive cell-state path help gradient flow?
    • It reduces reliance on repeated multiplicative transformations across time steps
    • O ṣe isodipupo ipinle nipasẹ nọmba nla ni igbesẹ kọọkan
    • O npa ipinle naa ni gbogbo igbesẹ
    • O di gbogbo awọn iwuwo lakoko ikẹkọ
  5. Iṣẹ imuṣiṣẹ wo ni awọn ẹnu-ọna LSTM nigbagbogbo lo lati ṣe bi awọn iyipada titan/pa a rirọ?
    • ReLU, ti njade 0 si ailopin
    • Softmax lori awọn fokabulari
    • Iṣẹ igbesẹ ti n jade ni deede 0 tabi 1
    • Sigmoid, ti njade 0 si 1
  6. Kilode ti Awọn Ayirapada fi rọpo awọn LSTM pupọ fun awoṣe ede ti o tobi?
    • Awọn ayirapada ko le mu ọrọ mu rara
    • Awọn LSTM ko nilo ikẹkọ
    • Awọn ayirapada ṣe afiwe ni ọna kan ki o gba aaye ti o gun-gun nipasẹ akiyesi
    • Ayirapada lo jina díẹ awọn nọmba ni lapapọ
  7. If an LSTM forget gate is near one and its input gate near zero, what happens to the cell state?
    • It largely retains the previous state
    • It is forced to zero
    • It becomes the raw input token
    • It must reverse the sequence
  8. Which gate controls how much of an LSTM cell’s stored state is exposed as output?
    • The forget gate alone
    • The output gate
    • The optimizer momentum term
    • The positional encoding gate