引導式測驗 · 硬 等級

貝葉斯深度學習測驗

探索貝葉斯網路、不確定權重、後驗近似、蒙地卡羅遺失以及保留資料的置信度校準。

相關引導路徑貝葉斯深度學習
問題 1 的 8

貝葉斯神經網路對待權重的方式與標準神經網路有何不同?

預覽每個問題8 問題

當您參加測驗時,就會出現答案和解釋。

  1. 貝葉斯神經網路對待權重的方式與標準神經網路有何不同?
    • 它將每個權重視為機率分佈而不是單一固定值
    • 它使用整數權重而不是小數
    • 它完全消除了重量
    • 它在所有層之間共享一個權重
  2. What can a Bayesian approach provide that a single fixed-weight network does not directly represent?
    • A guarantee that every confidence score is calibrated
    • A distribution of predictions that reflects uncertainty about model weights
    • A proof that the model will detect every unfamiliar input
    • A way to remove all noise from the data
  3. How can epistemic uncertainty change as relevant, representative training data are added?
    • It is guaranteed to reach zero with any added data
    • It is identical to label noise
    • It may decrease under the chosen model and data assumptions
    • It always increases linearly
  4. 為什麼需要變分推理或 MCMC 等近似方法?
    • 因為神經網路沒有損失函數
    • 因為GPU不能做加法
    • 因為權重必須是整數
    • 因為數百萬個權重的精確後驗在計算上是難以處理的
  5. 蒙特卡羅 dropout 是如何估計不確定性的?
    • 在測試時保持 dropout 處於活動狀態,並利用輸出的分佈多次運行網絡
    • 關閉輟學並延長訓練時間
    • 永久移除所有神經元
    • 用隨機噪音代替損失
  6. 變分推理將後驗估計重新定義為什麼樣的問題?
    • 一個排序問題
    • 最小化 KL 散度的最佳化問題(透過 ELBO)
    • 沒有目標的純隨機抽樣問題
    • 硬體設計問題
  7. How does Bayes’ rule relate a prior, likelihood, and posterior?
    • The posterior is the prior plus the likelihood, without normalization
    • It is the likelihood divided by the prior without normalization
    • The posterior is proportional to the prior times the likelihood
    • It is the fixed point estimate of the most likely weight vector
  8. How does MCMC approximate a Bayesian neural-network posterior?
    • By training a single point estimate with dropout disabled
    • By ranking model outputs with a fixed rule
    • By minimizing an ELBO with a simpler variational family
    • By drawing samples through a Markov chain whose stationary distribution targets the posterior