KDD 2019 | Policy Learning for Malaria Elimination

#1

Policy Learning for Malaria Elimination

Malaria is thought to have had the greatest disease burden throughout human history, while it continues to pose a significant but disproportionate global health burden. With 50% of the world’s population at risk of malaria infection though 90% of all cases contained in Sub Saharan Africa.

Through this KDD Cup Humanity RL track competition we are looking for participants to apply machine learning tools to determine novel solutions which could impact malaria policy in Sub Saharan Africa. Specifically, how should combinations of interventions which control the transmission, prevalence and health outcomes of malaria infection be distributed in a simulated human population.

This challenge has been framed as a Reinforcement Learning problem, participants are expected to submit high performing solutions to the provided sequential decision making task. For this competition actions receive stochastic and delayed rewards, which are resource constrained by both the monetary and computational costs associated with implementing and observing the impact of an action.

Submissions are encouraged from participants who may not have previous experience in reinforcement learning problems, reading through the materials we hope you may see the necessity for contributions to inform decision making for this complex real-world problem.

1 Like

#2

Competition Site : https://compete.hexagon-ml.com/practice/rl_competition/37/

0 Likes

#3

Introductory Reinforcement Learning Materials

All here are some free RL materials (Please feel free to add more to the list)

Free book - Sutton and Barto
http://incompleteideas.net/book/bookdraft2017nov5.pdf

0 Likes

#4

cannot access “https://nlmodelflask.eu-gb.mybluemix.net

0 Likes

#5

Hi @njmch03,

Per my understanding you don’t need to use this link anymore for the environment. The new code encapsulates details

See:
https://compete.hexagon-ml.com/practice/rl_competition/37/#getting_started_in_py

env = ChallengeEnvironment(experimentCount = 1000)
a = CustomAgent(env)
a.scoringFunction()
a.create_submissions("test.csv")

@oetbent @sekou

Can one of you provide more insights on this?

Taposh

0 Likes

#6

@taposhdr, Thanks a lot;

I tried “ChallengeEnvironment”, and cannot get any response;

See

It seems, “https://nlmodelflask.eu-gb.mybluemix.net” is called

Thanks again

0 Likes

#7

Hello!
Yes @taposhdr, that URI is being used to assess the policy, but it isn’t accessed it directly.
The CustomAgent class accesses this service through the self.environment.evaluateReward call.

0 Likes

#8

Thanks @sekou

@njmch03 what @sekou meant is that this is an api end-point and cannot be accessed directly. You should use the customAgent()

env = ChallengeEnvironment(experimentCount = 1000)
a = CustomAgent(env)
a.scoringFunction()
a.create_submissions("test.csv")
0 Likes

#9

Error: No module named ‘netsapi’ even after installing on google colab using following command:
!pip3 install git+https://github.com/slremy/netsapi --user --upgrade
I havn’t changed anything in example script just trying to run for first time!

Thank you.

0 Likes

#10

Can you please post screen shot of the error?

0 Likes

#11

0 Likes

#12

Can you please restart your runtime and try again?

1 Like

#13

I did, but same error.

0 Likes

#14

Hi @ashutoshaay26

I tried the same thing on colab notebook and it worked for me. See my gist below.

Regards,

Taposh

0 Likes

#15

Hi, Taposh. Same notebook I ran on colab and its showing error! Don’t know whats the problem?

0 Likes

#16

0 Likes

#17

Please try resetting your environment and also restart runtime if that does not work. I was able to see the same error but after reset and/or restart I am able to get it working again.

0 Likes

#18

Finally able to run it. Thanks for guidance and suggestions @super11 @taposhdr

2 Likes

#19

Very cool !! @ashutoshaay26

1 Like

#20

But everytime haveto reset and run again on colab! Whats the issue?

0 Likes