Community Digest

Top new questions this week:

How to handle the dead agent in multi-agent environment?

I try to implement deep reinforcement learning on a defender-vs-attacker problem, where agents can be destroyed by enemies. I am coding both the environment and the RL algorithm. The agent can observe ...

deep-rl  
user avatar asked by zhixin Score of 1
user avatar answered by Neil Slater Score of 2

Why solely a one-step-lookahead in value/policy-iteration?

In value iteration and policy iteration we solely consider a one-step-lookahead where the lookahead is from the previous iteraiton and therefore need to sweep over all states and iterate this ...

reinforcement-learning markov-decision-process value-iteration model-based-methods dynamic-programming  
user avatar asked by hugh Score of 1

How to plot average return vs step figures in reinforcement learning?

I know return is total discounted reward of an episode,so it is easy to plot return vs. episode.But in many papers,they provide figures of average return vs step,like this: Based on my knowledge,...

reinforcement-learning  
user avatar asked by waylone Score of 1
user avatar answered by Alberto Score of 1

Data preparation for NLP model

I have data from our ticketing system. Currently using OpenNLP to create different models. For simplicity I have a 10k ticket's text as category final queue of the ticket. My questions: Is it ...

machine-learning natural-language-processing language-model  
user avatar asked by Milkmaid Score of 1
user avatar answered by Alberto Score of 0

Search recall optimization - what appropriate loss function to use?

I am studying machine learning and wanted to work on a project of my own so that I have better chances after graduating college. I'm studying the application of ML to improve searches using a toy ...

classification objective-functions search  
user avatar asked by user9343456 Score of 1

UCB, Thompson sampling etc seems myopic/greedy for bandits?

When considering multi-armed bandits in different formats, UCB, $\epsilon$-greedy, thompson sampling etc seems so greedy/myopic in the sense that it solely considers reward for the current timestep. ...

reinforcement-learning markov-decision-process multi-armed-bandits upper-confidence-bound thompson-sampling  
user avatar asked by hugh Score of 1
user avatar answered by maxy Score of 1

How does Alpha Go Zero MCTS work in parallel?

I am trying to better understand the article "Mastering the Game of Go without Human Knowledge" (link) and I'm confused about the parallel implementation of Monte-Carlo-Tree-Search. On page ...

monte-carlo-tree-search alphago-zero  
user avatar asked by martinkunev Score of 1
user avatar answered by DSarkar Score of 2

Greatest hits from previous weeks:

1 hidden layer with 1000 neurons vs. 10 hidden layers with 100 neurons

These types of questions may be problem-dependent, but I have tried to find research that addresses the question whether the number of hidden layers and their size (number of neurons in each layer) ...

neural-networks  
user avatar asked by Stephen Johnson Score of 17
user avatar answered by Thomas Wagenaar Score of 14

What is convergence in machine learning?

I came across this answer on Quora, but it was pretty sparse. I'm looking for specific meanings in the context of machine learning, but also mathematical and economic notions of the term in general.

neural-networks machine-learning terminology convergence  
user avatar asked by DukeZhou Score of 8
user avatar answered by ashenoy Score of 6

OpenAI: What is the difference between model "gpt-3.5-turbo" and "gpt-3.5-turbo-0301"?

I have performed an API call to OpenAI's endpoint https://api.openai.com/v1/models . The endpoint lists the currently available engines, and provides basic information about each one such as the owner ...

open-ai chatgpt large-language-models  
user avatar asked by knb Score of 2
user avatar answered by jamiecropley Score of 3

What are the limitations of the hill climbing algorithm and how to overcome them?

What are the limitations of the hill climbing algorithm? How can we overcome these limitations?

algorithm search optimization problem-solving hill-climbing  
user avatar asked by Abbas Ali Score of 10
user avatar answered by Ugnes Score of 6

How can I predict the next number in a non-obvious sequence?

I've got an array of integers ranging from -3 to +3. Example: [1, 3, -2, 0, 0, 1] The array has no obvious pattern since it represents bipolar disorder mood swings. What is the most suitable approach ...

machine-learning recurrent-neural-networks prediction sequence-modeling  
user avatar asked by ZenBerry Score of 8
user avatar answered by serali Score of 9

Why is Sanskrit the best language for AI?

According to NASA scientist Rick Briggs, Sanskrit is the best language for AI. I want to know how Sanskrit is useful. What's the problem with other languages? Are they really using Sanskrit in AI ...

comparison programming-languages nasa  
user avatar asked by Rahul Score of 16
user avatar answered by Christian Westbrook Score of 16

Why do CNN's sometimes make highly confident mistakes, and how can one combat this problem?

I trained a simple CNN on the MNIST database of handwritten digits to 99% accuracy. I'm feeding in a bunch of handwritten digits, and non-digits from a document. I want the CNN to report errors, so I ...

convolutional-neural-networks tensorflow keras bayesian-deep-learning uncertainty-quantification  
user avatar asked by Alexander Soare Score of 37

Can you answer these questions?

Why policy gradient theorem has two different forms?

I have been studying policy gradients recently but found different expositions from different sources, which greatly confused me. From the book "Reinforcement Learning: an Introduction (Sutton &...

reinforcement-learning deep-learning deep-rl policy-gradients policy-gradient-theorem  
user avatar asked by Yuxiang Wei Score of 1

How does diffusion model (DDPM) ensures novel generated samples?

I am trying to understand the theoretical aspect of the denoising diffusion model. There we try to destroy the initial image x_0 through a chain of forward process and then learn a backward diffusion ...

deep-learning generative-model diffusion-models  
user avatar asked by Formal_this Score of 1

DQN Loss function - doubt about stochastic approximation

In Deep Q Learning algorithm the convergence is generally achieved using smart tricks like the target network and the replay buffer. However there is one thing which is not clear to me. When the Q ...

dqn stochastic-gradient-descent  
user avatar asked by Ftoso91 Score of 1
user avatar answered by Alberto Score of 0
You're receiving this message because you subscribed to the Artificial Intelligence community digest.
Unsubscribe from this community digest       Edit email settings       Leave feedback       Privacy
Stack Overflow

Stack Overflow, 14 Wall Street, 20th Floor, New York, NY 10005

<3