Главная
Study mode:
on
1
Start
2
Deep Reinforcement Learning of Marked Temporal Point Processes
3
Many discrete events in continuous time
4
Variety of processes behind these events
5
Example I: Information propagation
6
Example II: Knowledge creation
7
Aren't these event traces just time series?
8
What are marked temporal point processes?
9
What can MTPPs model?
10
What can MTPPs model: when-to-post
11
What can MTPPs model: spaced-repetition
12
How to optimize Agent's policy?
13
Optimizing Agent's policy using RL
14
Outline
15
Representing Marks and Times of MTPPs
16
How to represent MTPPs: timing of events
17
How to represent MTPPs: marks of events
18
How to represent MTPPs: summary
19
Reinforcement Learning: Setup
20
Reinforcement Learning: Discrete time
21
Reinforcement Learning: Continuous time
22
RL with entire history as state
23
RL state: embedding marks
24
RL state: embedding source of event
25
RL state in parametrization of the policy
26
RL with Asynchronous Feedback
27
RL problem with MTPPs: summary
28
Policy optimization problem
29
Existing approaches have limitations
30
Policy Gradient method can be used!
31
Policy Gradient: Example iteration
32
Spaced repetition: Problem setup
33
Spaced repetition to smart repetition
34
When-to-post: Problem setup
35
When to post with unknown priorities
36
When to post with baselines
37
Deep Reinforcement Learning for Marked Temporal Point Processes
38
Thank you!!
Description:
Explore deep reinforcement learning techniques for marked temporal point processes in this 36-minute conference talk. Delve into the modeling of discrete events in continuous time, examining examples like information propagation and knowledge creation. Learn how to represent the timing and marks of events in marked temporal point processes (MTPPs) and understand their applications in scenarios such as optimizing when-to-post strategies and spaced repetition learning. Discover the reinforcement learning setup for continuous time processes, including state representation and policy optimization using policy gradient methods. Gain insights into solving real-world problems like smart repetition scheduling and strategic social media posting through the application of deep reinforcement learning to MTPPs.

Deep Reinforcement Learning of Marked Temporal Point Processes

International Centre for Theoretical Sciences
Add to list
#Computer Science #Machine Learning #Reinforcement Learning #Deep Reinforcement Learning #Policy Gradient Methods
0:00 / 0:00
1 of 38

Start