3 votes
Accepted

Intuition behind $1-\gamma$ and $\frac{1}{1-\gamma}$ for calculating discounted future state distribution and discounted reward

Question 1 The taylor expansion of $\frac{1}{1-\gamma}$ at $\gamma= 0$ is as follows $$\frac{1}{1-\gamma} = 1 + \gamma + \gamma^2 + \dots$$ When you multiply by $1-\gamma$ you get $$ 1 = (1-\gamma)(1 +...
2 votes

How do we design a neural network such that the $L_1$ norm of the outputs is less than or equal to 1?

Penalty (barrier function) is perfectly valid and simplest method for simplex type constraint (L1 norm is simplex constraint on absolute values). Any type of barrier function may work, logarithmic, ...
2 votes

How does one make a neural network learn the training data while also forcing it to represent some known structure?

Extending @mirror2image's comment, if you have a certain metric that allows you to measure how close the intended layer is to a low pass filter (something that compares its output with what a low pass ...
2 votes
Accepted

How to use DQN when the action space can be different at different time steps?

There are two relevant neural network designs for DQN: Model q function directly $Q(s,a): \mathcal{S} \times \mathcal{A} \rightarrow \mathbb{R}$, so neural network has concatenated input of state and ...
  • 24.7k
1 vote

How can we design the mutation and crossover operations when the order of the genes in the chromosomes matters?

If I understood correctly, your problem is about finding the optimal way to execute a series of tasks in order to maximize the results, using Genetic Algorithms. In few words, you're trying to ...
1 vote

Given a list of integers $\{c_1, \dots, c_N \}$, how do I find an integer $D$ that minimizes the sum of remainders $\sum_i c_i \text{ mod } D$?

Genetic Algorithm is not the best approach here: GA is a stochastic methods and therefore will never guarantee the best possible solution. Your solution (GA individual) is modeled as a simple integer....
1 vote

How to handle infeasibility caused due to crossover and mutation in genetic algorithm for optimization?

You have two broad categories of options, prevention and repair. Prevention means defining a crossover and mutation operator that try to be more intelligent about respecting the constraints. Suppose ...
  • 446
1 vote

How to use a VAE to reconstruct an image starting from an initial image instead of starting from a random vector?

The thing is, the decoder samples from a latent mu and sigma, so you cant sample from a raw image directly. But if you’re trying to put a random image into the encoder of a trained VAE to match it to ...
  • 111

Only top scored, non community-wiki answers of a minimum length are eligible