It's my understanding we have the following risk scores available in LB
CMS-HCC risk score
HHS-HCC risk score
Johns Hopkins risk score (ACG)
ATI risk score
Are there any other risk scores available? Is there only one risk score available from the Johns Hopkins methodology? A client has said their understanding was that there were multiple scores available for them.
In order to calculate ATI risk score, does LB look back at 12 months of data or more?
Related
I'm wondering what is the limit of Bing Distance Matrix API. From the docs here said 2,500 origins-destination pairs. Is it the maximum request per single request or the cumulative monthly limit? Another thing that confused me is about their billable transaction. What is it about the billable transaction?
I did try reading the docs and implementing the API. I'm a little afraid to try a 50x50 matrix as I will get charged.
That limit is per request. There is no monthly limit.
Non-billable transactions are tracked, but don't count towards the free limits (or paid for transaction credits). Billable transactions eat up the free credits and potentially into purchases transaction credits, if you have any.
Note that with Bing maps you have to purchase transactions before you can use them. So, if you don't have a license today, you can at most only max out the free limits. Once you hit the free limits, the service will stop working for your account until you decide to get a paid license, or until a new year has come around. So, you don't need to worry about a surprise bill appearing in this case.
In Bing Maps, every 4 cells in a matrix (origin/destination pair) generates 1 transaction. So, a 50x50 matrix would have 2,500 cells, and thus use 625 transactions. The Bing Maps free limits is 125,000 transactions a year. So, you could in theory run request a 50x50 matrix every other day and stay under the free yearly limit.
Currently I have a stock of Grade 1 learners. The stock is arrayed so that Gr1_learners[..] = {10,10} means that there are 10 high performing learners and 10 low performing learners in this stock. These learners progress to Grade 2 so that Gr2_learners[..] = {20,0} are the first graders having progressed after one year and some intervention improved all the learners to high performing.
My idea is to create learners as agents and place them in the stocks. Then Gr1_learners = 20 is a stock with 20 agents, half in state high_performing, and half in state low_performing. They progress, a state change is triggered so that Gr2_learners = 20 is then a stock with 20 agents, all in state high_performing and none in state low_performing.
Is this possible?
Yes, by not using stocks at all. You are probably familiar with SD modelling and now see the benefit of agent-based modelling.
Go all in and model your agents. Instead of stocks, you use agent populations and make your agents change states.
See the agent example models, especially the SIR one (there is an SD version and an ABM version, great to see how you can model any SD model with agents)
But you cannot have agents inside SD stocks, they are not actual stocks but just mathematical equations :)
Currently I'm building my monitoring services for my e-commerce Server, which mostly focus on CPU/RAM usage. It's likely Anomaly Detection on Timeseries data.
My approach is building LSTM Neural Network to predict next CPU/RAM value on chart trending and compare with STD (standard deviation) value multiply with some number (currently is 10)
But in real life conditions, it depends on many differents conditions, such as:
1- Maintainance Time (in this time "anomaly" is not "anomaly")
2- Sales time in day-off events, holidays, etc., RAM/CPU usages increase is normal, of courses
3- If percentages of CPU/RAM decrement are the same over 3 observations: 5 mins, 10 mins & 15 mins -> Anomaly. But if 5 mins decreased 50%, but 10 mins it didn't decrease too much (-5% ~ +5%) -> Not an "anomaly".
Currently I detect anomaly on formular likes this:
isAlert = (Diff5m >= 10 && Diff10m >= 15 && Diff30m >= 40)
where Diff is Different Percentage in Absolute value.
Unfortunately I don't save my "pure" data for building neural network, for example, when it detects anomaly, I modified that it is not an anomaly anymore.
I would like to add some attributes to my input for model, such as isMaintenance, isPromotion, isHoliday, etc. but sometimes it leads to overfitting.
I also want to my NN can adjust baseline over the time, for example, when my Service is more popular, etc.
There are any hints on these aims?
Thanks
I would say that an anomaly is an unusual outcome, i.e. a outcome that's not expected given the inputs. As you've figured out, there are a few variables that are expected to influence CPU and RAM usage. So why not feed those to the network? That's the whole point of Machine Learning. Your network will make a prediction of CPU usage, taking into account the sales volume, whether there is (or was) a maintenance window, etc.
Note that you probably don't need an isPromotion input if you include actual sales volumes. The former is a discrete input, and only captures a fraction of the information present in the totalSales input
Machine Learning definitely needs data. If you threw that away, you'll have to restart capturing it. As for adjusting the baseline, you can achieve that by overweighting recent input data.
I am making an AI for a zero-sum 4-player board game. It's actually not zero-sum (the 4 players will "die" when they lose all their lives, so there will be a player who died first, second, third and a player who survived. However, I am telling the AI that only surviving counts as a win and anything else is a lose) After some research, I figured I would use a minimax algorithm in combination with a heuristic function. I came across this question and decided to do the same as the OP of that question - write an evolutionary algorithm that gives me the best weights.
However, my heuristic function is different from the one the OP of that question had. Mine takes 9 weights and is a lot slower, so I can't let the agents play 1000 games (takes too much time) or breed them with the crossover method (how do I do a crossover with 9 weights?).
So I decided to come up with my own method of determining fitness and breeding. And this question is only about the fitness function.
Here are my attempts at this.
First Attempt
For each agent A in a randomly generated population of 50 agents, select 3 more agents from the population (with replacement but not the same agent as A itself) and let the 4 agents play a game where A is the first player. Select another 3 and play a game where A is the second player, and so on. For each of these 4 games, if A died first, its fitness does not change. If A died second, its fitness is increased by 1. If it died third, its fitness is increased by 2. If it survived, its fitness is increased by 3. Therefore, I concluded that the highest fitness one can get is 12 (surviving/wining all 4 games -> 3 + 3 + 3 + 3).
I ran this for 7 generations and starting from the first generation, the highest fitness is as high as 10. And I calculated the average fitness of the top 10 agents, but the average didn't increase a bit throughout the 7 generations. It even decreased a little.
I think the reason why this didn't work is because there's gotta be a few agents that got lucky and got some poor performing agents as its opponents.
Second Attempt
The game setups are the same as my first attempt but instead of measuring the results of each game, I decided to measure how many moves did that agent make before it died.
After 7 generations the average fitness of top 10 does increase but still not increasing as much as I think it should.
I think the reason why this failed is that the game is finite, so there is a finite number of moves you can make before you die and the top performing agents pretty much reached that limit. There is no room for growth. Another reason is that the fitness of the player who survived and the fitness of the player who died third differs little.
What I want
From my understanding of EAs (correct me if I'm wrong), the average fitness should increase and the top performing individual's fitness should not decrease over time.
My two attempts failed at both of these. Since the opponents are randomly selected, the top performing agent in generation 1 might get stronger opponents in the next generation, and thus its fitness decreases.
Notes
In my attempts, the agents play 200 games each generation and each generation takes up to 3 hours, so I don't want to let them play too many games.
How can I write a fitness function like this?
Seven generations doesn't seem like nearly enough to get a useful result. Especially for a game, I would expect something like 200+ generations to be more realistic. You could do a number of things:
Implement elitism in order to ensure the survival of the best individual(s).
The strength of evolution stems from repeated mutation and crossover, so I'd recommend letting the agents play only a few games per generation (say, 5 ~ 10), at least at the beginning, and then evolve the population. You might even want to do only one game per generation.
In this regard, you could adopt a continuous evolution strategy. What this means is that as soon as an agent dies, they are subjected to mutation, and as soon as an agent wins, they can produce offspring. Or any combination of the two. The point is that the tournament is ongoing, everyone can play against anyone else. This is a little more "organic" in the sense that it does not have strictly defined generations, but it should speed up the process (especially if you can parallelise the evaluation).
I hope that helps. The accepted answer in the post you referenced has a good suggestion about the way you could implement crossover.
I want to be able to rank users based on how quick they have completed each level. I want this to be an overall leaderboard I.e. shortest overall time for all levels.
The problem here is that for each level completed the totally completion time goes up. But I want to ensure that the leaderboard takes that into account so that a user having completed 10 levels will rank more highly than someone with only 1 completed level.
How can I create some kind of score based on this?
Before submitting the time to leader board.
You could perform a modulation on the total time by the number of levels completed, then for each level completed reduce it by a set amount so people who complete all levels with the same average time will score better then people with the same average time but with fewer levels.
My Preferred Method:
Or you could express it with a score value.
level complete = 1,000.
Each level has a set time limit bonus, the longer you take the less bonus u get.
eg
I Complete the level in 102 secs Goal time is 120 secs
I get 1,000 points for completion and 1,500 points for each second
that i beat the Goal time for.
This way i will get 1,000 + (18* 1,500) = 28,000 points
Next guy completes in 100 secs
He Gets 1,000 + (20*1,500) = 31,000 points
I suggest adding a default amount of time to the total for each incomplete level. So, say, if a player beats a new level in 3 minutes, that replaces a 10 minute placeholder time, and they 'save' 7 minutes from the total.
Without that kind of trick, the iPhone has no provision for multi-factor rankings.
Leaderboard scores in GameKit have to be expressed as a single number (see this section of the GameKit Programming Guide), so that won't be possible.
Your best bet would be to just have a completion time leaderboard for people who have completed all the levels, and maybe another leaderboard (or a few) for people who have completed a smaller number of levels.