DRL
- 95 שאלות
- 1 תגובות
- 0% הושלמו
- equalizer סטטיסטיקות
- share שתף
מנהלים:
Discuss, Learn and be Happy דיון בשאלות
help
brightness_4
brightness_7
format_textdirection_r_to_l
format_textdirection_l_to_r
b) Experience replay is generally less effective in stochastic environments compared to deterministic environments due to the increased variance in sampled transitions
1
| done |
מיין לפי
16. [General] Which of the following statements regarding actor-critic methods is correct:
1
| done | ||
מיין לפי
b) In the approach “Curiosity-driven Exploration by Self-supervised Prediction”, in environments where all states are equally novel or predictable, this curiosity-driven approach would behave identically to a random exploration strategy.
1
| mood |
מיין לפי
19. [General] Which of the following statements regarding policy/value iteration is correct:
1
| done | ||
מיין לפי
b) The jointly-learned state-action embedding approach allows for more efficient exploration by leveraging similarities between actions in the embedding space
1
| done |
מיין לפי