Trjtdtk

Can disgust be a key component of horror?

Why can Carol Danvers change her suit colours in the first place?

Is this toilet slogan correct usage of the English language?

Can a College of Swords bard use a Blade Flourish option on an opportunity attack provoked by their own Dissonant Whispers spell?

How does a computer interpret real numbers?

Mimic lecturing on blackboard, facing audience

How do you make your own symbol when Detexify fails?

Picking the different solutions to the time independent Schrodinger eqaution

Hero deduces identity of a killer

Do the primes contain an infinite almost arithmetic progression?

How much character growth crosses the line into breaking the character

When were female captains banned from Starfleet?

How could a planet have erratic days?

How should I respond when I lied about my education and the company finds out through background check?

Why is so much work done on numerical verification of the Riemann Hypothesis?

Temporarily disable WLAN internet access for children, but allow it for adults

Creepy dinosaur pc game identification

Why "had" in "[something] we would have made had we used [something]"?

Can I still be respawned if I die by falling off the map?

What is going on with 'gets(stdin)' on the site coderbyte?

Plot of a tornado-shaped surface

Why is it that I can sometimes guess the next note?

Quoting Keynes in a lecture

What is the highest possible scrabble score for placing a single tile

Modeling a Neural Network for a Turn-based Strategy (TBS) game

Neural Network Hidden Neuron Selection StrategyHow to teach neural network a policy for a board game using reinforcement learning?Using Neural Networks To Predict SetsDeep neural net modelling strategyNeural Network Timeseries Modeling with Predictor VariablesHow to normalize data for Neural Network and Decision ForestHow to prevent a neural network from choosing the 'easiest' solutionWhat type of NN should be used for a turn-based game (with direct access to data)?What kind of neural network would work best for loosely-defined data, like video game RAM?Hindsight experience replay: strategy for sampling goals

0

$begingroup$

I'm modeling a neural network for a turn-based strategy game that involves purchase items every round. After the purchase phase, a combat phase is automatically executed (the winner gets more gold and the loser, consequently, minus gold). It is possible to buy more than one item, if there is available resource (gold) and the maximum number of items that it is possible to have is 8. In addition, it stands out the fact that the available items can be the same or different every round from the previous one. The main purpose of the neural network is to predict which items to buy.

What I thought to do is to create some features in order to feed the input layer of the network in the following manner: available gold; purchased items so far (max 8); available items for purchase (max 5); price of the available items for purchase. As output from the network would be 5 neurons, each representing the available items for purchase. So I am afraid of the following situation: I have 2 gold available and the neural network returns me that I should buy an item that costs 1 gold and another that costs 2, which would not be allowed. How could I deal with this? Has anyone ever dealt with a similar scenario?

Thank you very much!!

neural-network deep-learning reinforcement-learning research game

share|improve this question

asked 5 hours ago

Matheus PrandiniMatheus Prandini

504

$endgroup$

add a comment | 

0

$begingroup$

I'm modeling a neural network for a turn-based strategy game that involves purchase items every round. After the purchase phase, a combat phase is automatically executed (the winner gets more gold and the loser, consequently, minus gold). It is possible to buy more than one item, if there is available resource (gold) and the maximum number of items that it is possible to have is 8. In addition, it stands out the fact that the available items can be the same or different every round from the previous one. The main purpose of the neural network is to predict which items to buy.

What I thought to do is to create some features in order to feed the input layer of the network in the following manner: available gold; purchased items so far (max 8); available items for purchase (max 5); price of the available items for purchase. As output from the network would be 5 neurons, each representing the available items for purchase. So I am afraid of the following situation: I have 2 gold available and the neural network returns me that I should buy an item that costs 1 gold and another that costs 2, which would not be allowed. How could I deal with this? Has anyone ever dealt with a similar scenario?

Thank you very much!!

neural-network deep-learning reinforcement-learning research game

share|improve this question

asked 5 hours ago

Matheus PrandiniMatheus Prandini

504

$endgroup$

add a comment | 

$begingroup$

I'm modeling a neural network for a turn-based strategy game that involves purchase items every round. After the purchase phase, a combat phase is automatically executed (the winner gets more gold and the loser, consequently, minus gold). It is possible to buy more than one item, if there is available resource (gold) and the maximum number of items that it is possible to have is 8. In addition, it stands out the fact that the available items can be the same or different every round from the previous one. The main purpose of the neural network is to predict which items to buy.

What I thought to do is to create some features in order to feed the input layer of the network in the following manner: available gold; purchased items so far (max 8); available items for purchase (max 5); price of the available items for purchase. As output from the network would be 5 neurons, each representing the available items for purchase. So I am afraid of the following situation: I have 2 gold available and the neural network returns me that I should buy an item that costs 1 gold and another that costs 2, which would not be allowed. How could I deal with this? Has anyone ever dealt with a similar scenario?

Thank you very much!!

neural-network deep-learning reinforcement-learning research game

share|improve this question

asked 5 hours ago

Matheus PrandiniMatheus Prandini

504

$endgroup$

share|improve this question

asked 5 hours ago

Matheus PrandiniMatheus Prandini

504

share|improve this question

asked 5 hours ago

Matheus PrandiniMatheus Prandini

504

asked 5 hours ago

Matheus PrandiniMatheus Prandini

504

asked 5 hours ago

Matheus PrandiniMatheus Prandini

504