Haitham Bou Ammar

RL team leader @Huawei R&D UK & UCL H. Assistant Prof.

Haitham Bou Ammar RL team leader @Huawei R&D UK & UCL H. Assistant Prof.

发布于2022-03-07 02:52:21
A 4 step proof that value baselines don't affect policy grads in #RL😀Just the log-trick & Fubini gets u there!
赞 8

评论 4

浏览 1509

课程学习
Haitham Bou Ammar RL team leader @Huawei R&D UK & UCL H. Assistant Prof.

发布于2022-03-02 22:47:19

Iterated Law of Expectation Consice Proof (6 simple steps)
When deriving the Belman equations, we needed the iterated law of expectations. Rather than believing me, have a look at this 6-...

赞 3

评论 1

浏览 1193

课程学习
Haitham Bou Ammar RL team leader @Huawei R&D UK & UCL H. Assistant Prof.

发布于2022-03-02 05:44:26
ELBO in 5 simple steps starting direclty from Bayes Rule!
赞 3

评论

浏览 1119

课程学习
Haitham Bou Ammar RL team leader @Huawei R&D UK & UCL H. Assistant Prof.

发布于2022-02-28 22:26:57

(RL 正在改变世界！贝尔曼方程是#RL 的核心。这是一个 3 (3.5:P) 步骤的简明证明)

赞 6

评论 2

浏览 1294

课程学习