Skip to main content

Questions tagged [artificial-intelligence]

Questions about design and properties of agents that act in a dynamic environment and make decisions towards some goal without user control.

Filter by
Sorted by
Tagged with
0 votes
1 answer
22 views

Forming Sentences

I have a bunch of transcripts from online videos but I suspect the transcripts aren't formatted well. There are no punctuations, sentences are broken abruptly, some words aren't complete (for example ...
1 vote
2 answers
93 views

I'm quite confused about the convergence of pix2pix cGAN, how can we tell whether my model is well trained?

Usually we will see decreasing loss for regular CNN model, but for GAN, it seems to be not that simple. How can we tell whether my model is well trained?
1 vote
2 answers
438 views

TOPS trillion operations per second to Tokens per second

A lot of AI hardware coming out lately has its performance mentioned in TOPS i.e trillion operations per second. Does anyone have an Idea how to estimate the llm performance on such hardware in tokens ...
1 vote
0 answers
24 views

Simplest way to incorporate edge types into self-attention (in a graph transformer)?

In the GT (Graph Transformer) model presented by Dwivedi & Bresson in "A Generalization of Transformer Networks to Graphs", the following equations are used to update node and edge ...
1 vote
1 answer
415 views

Formal definition of loss surface of multi-layered networks

Let $\mathcal{L}$ be a loss function associated with a multi-layered neural network. So it seems almost everyone in AI/ML community is interested in the Hessian $H=\partial^2 \mathcal{L}$ of $\...
0 votes
1 answer
61 views

Main subjects to learn Artificial Intelligence in CS

In my PhD, I will work with ML models. However, I will only use ready-made models as a tool, but I want to delve deeper into Artificial Intelligence not just to use ready-made models, but to ...
1 vote
1 answer
140 views

Optimality of A* algorithm

In the book Artificial Intelligence a Modern Approach 4th edition, the author claims that the A* algorithm is cost-optimal if the heuristic function is ...
0 votes
1 answer
84 views

Performing semantic segmentation on images on word documents

At the moment, I'm making a comparative analysis between two different neural networks for a class. The problem they are both trying to solve is detecting tables on documents. Below is an example of ...
1 vote
1 answer
85 views

Logical Consequence - Equivalent Assertions

I have the following slide in my notes and I'm having trouble understanding how the three assertions are equivalent. I understand to a degree how the 2nd and 3rd assertions are equivalent, but the ...
0 votes
1 answer
105 views

Confused on Variational AutoEncoder

I'm a bit confused on how Variational Autoencoders are trained. In particular I'm confused on how the latent variable is generated for each input. My questions are as follows: When running ...
1 vote
2 answers
83 views

How much do llms know?

Can we assume that a 7gb llm model knows 7gb of text info? (by 7gb i mean a 7 billion parameter model at Q8 quantization) or a (1.75 billion parameter model at full fp32 precision) For example, a 70,...
1 vote
2 answers
66 views

Gödel's theorem and machines' power

I was studying AI and when a question came to my mind. I know that one of the objections to the possibility of a thinking machine examined by Turing is the so called mathematical objection, ...
0 votes
1 answer
186 views

BFS on a graph and BFS on a tree

I found the following question in my book and I have no clue on what the answer should be: What is the condition on search graph so that BFS Algorithm for graph and BFS Algorithm for tree generate ...
1 vote
1 answer
17 views

linear relationship between the log-odds and the features

In this post I asked about why the sigmoid/softmax function was used in classification: Binary Classification- Non-Differentiable Loss Function But I have a followup question: We're assuming that the ...
0 votes
1 answer
38 views

Question about data to train an machine learning IA

I'm trying to create a machine learning program to predict the winner of UFC fights. I want to train my AI with data from previous matches, but I have encountered a problem. I wonder if the algorithm ...
1 vote
1 answer
24 views

Binary Classification- Non-Differentiable Loss Function

For binary classification using linear regression, we pass the output z of the linear regression through the sigmoid function so that if the linear regression takes an input x which should be ...
1 vote
1 answer
111 views

How to think about heuristics

I have a game about drone delivery company. The game consists of a rectangular map, which contains tiles that cannot be passed. I can control a certain amount of drones, each one of them starts at a ...
0 votes
0 answers
12 views

Predict output using image and text variables as input

I need to implement a solution where I will have an image and other text variables as temperature, humidity and precipitacion. I wish to preview the inflow as output based on these inputs. I suspect ...
0 votes
0 answers
10 views

Partial derivatives wrt input and weights in CNNs

I'm working on understanding the backpropagation process in convolutional neural networks (CNNs) and am having some trouble comprehending how to calculate the partial derivatives of the output of a ...
0 votes
0 answers
119 views

Is deepfake detection viable?

I'm thinking of doing a project on deepfake detection, but I'm not entirely sure if it is viable. Based on my understanding, how it works is that deepfake generation programs have a generative and ...
2 votes
1 answer
359 views

Potential of Artificial Intelligence to improve algorithm for PI

Recently AI (alphazero) was used to improve practical matrix multiplications by around 10 to 20 percent. Alphageometry and ramanujan machines all came into existence in the recent years as well. Seems ...
2 votes
1 answer
117 views

How do I make an A.I. reward system?

This question is because I think it would be fun to make an A.I. Any programming language is acceptable. I am wondering how do I make a simple program that can associate certain things with points and ...
0 votes
0 answers
27 views

Explanation for the expression of positional encoding in NeRFs

I was reading the NeRF paper recently (https://arxiv.org/pdf/2003.08934.pdf), and under the positional encoding section, I see that the authors propose usage of the following function for transforming ...
0 votes
0 answers
16 views

Best solution architecture for: Intelligent chat to retrieve information from specific sources

I am trying to design best solution for a chat application based on LLM model which has set of functions and based on user input can retrieve the information. I would like to ask you for some ...
0 votes
2 answers
63 views

Trying to implement BFS and I am stuck

I am trying to write down a code which would blindly search for a condition using breadth first search.I have been thinking of it for quite some time and I cant figure out how to continue. On the one ...
0 votes
1 answer
51 views

Is it possible that an AI bot behave as a human being in the long term?

Is it possible that an AI bot behave as a human being in the long term? I would like to know what people think but according to me it is not possible. One example is - Many times humans know what is ...
0 votes
0 answers
49 views

Reading and answering a text question from a image?

I wanted to see what models would be best suited for providing a image (see the link below) and having the model perform a image to text (OCR of some kind) and laying out the contents of the image in ...
0 votes
1 answer
135 views

Why can't humans translate all of their abilities into an algorithm? Will that hold true in the future?

I know that an algorithm can't decide whether another algorithm halts on an input or not (a Turing machine can't decide whether another Turing machine will halt on an input). But I, as an human, can: ...
10 votes
1 answer
9k views

The meaning of discount factor on reinforcement learning

After reading of the google deepmind achievements on Atari's games, I am trying to understand the q-learning and q-networks, but I am little bit confused. The confusion arise in the concept of the ...
21 votes
3 answers
20k views

When should I learn artificial intelligence? [closed]

Straight to the point: I would really like to learn AI. But I want some advice from experienced CS guys as to when I should jump into Artificial Intelligence. What prerequisites are needed in order ...
-3 votes
1 answer
278 views

If Q* can break encryption would that prove P=NP?

At 12:11 in this video the creator talks about unverified rumors the Q* algorithm can break AES-192 encryption. If this is true, would this mean P = NP?
-2 votes
1 answer
49 views

Objective function

I want to learn how to write objective functions for any problem or real-life action(not mathematical problems) can anyone suggest a good source? On online I am mostly finding objective functions for ...
0 votes
0 answers
27 views

Manhattan distance always less node expansion than misplaced tiles heuristic?

I created a 8-puzzle search solver using BFS, A* with manhattan distance, and A* with misplaced tiles. I generated data that said that for a particular random board, misplaced tiles did less node ...
1 vote
1 answer
41 views

Are there already Neural Network AIs that can play object-finding games?

Neural networks are good at image analysis. They are also being used in video games which could outperform professional human players. I wonder if they can also play a purely object-finding game where ...
0 votes
0 answers
40 views

Incompresensible about $w$ $x$+ $b$ = 0

I don't actually understand the meaning of $w$$x$ + $b$ = 0 when it is defined in support vector machine. In my own understanding, in order for the equation to be true, the hyperplane would always ...
0 votes
2 answers
410 views

Support Vector Machines vs K-Nearest Neighbors

Let's say we have trained a Support Vector Machine with a Gaussian Kernel. When we feed our model an example, it classifies it based on its similarity to landmarks (distance to examples in our ...
0 votes
0 answers
33 views

Finding an algorithm EF[1,1] and PO division for more than two agents

From this research paper I want to write an algorithm for finding envy-freeness(EF) and Pareto optimality(PO) division for more than two agents. We consider the problem of fairly and efficiently ...
0 votes
2 answers
180 views

a program discovering himself how to solve propositional calculus

it is well-known that propositional logic problems such as $$ (p\leftrightarrow q) \lor r \quad\overset{?}{\vdash}\quad (((p\lor q)\to(p\land q)) \land \lnot r)\lor r$$ can be simply solved by ...
15 votes
2 answers
21k views

Initial temperature in simulated annealing algorithm

I've done some testing of different initial temperatures in my simulating annealing algorithm and noticed the starting temperature has an affect on the performance of the algorithm. Is there any way ...
2 votes
0 answers
90 views

How does this star1 variation work? (star1 is a stochastic (dice) algorithm for artificial intelligence in adversarial games)

I don't understand how the star1 variation mentioned in the paper is Sparse Sampling for Adversarial different than the usual star1 algorithm? (Taken from the bottom of page 7). This variant can be ...
-1 votes
1 answer
62 views

Convolutional neural network for non-image data

I am working on a project for the prediction of preeclampsia, and one of the algorithms to be used is the convolutional neural network. I am a beginner in machine learning, and I am struggling to ...
1 vote
0 answers
119 views

At what precision are GPT's parameters stored?

OpenAI's technical paper on GPT-3 says that GPT-3 has 175 billion parameters. Several sites (here, here) claim that these parameters are stored as single-precision floating-point numbers and so ...
1 vote
1 answer
321 views

Why GPT model is a higher order hidden markov model

I have read the GPT-1 paper, and my understanding is that it works as follows: $U$ is input tokens, $h_0=UW_e+W_p$, $h_i=\text{transformer_block}(h_{i-1})$ and the output is a probability vector $P(u)=...
0 votes
1 answer
49 views

Is there an AI that can play "One of these things is not like the other" (Sesame Street)?

Refers to: https://en.wikipedia.org/wiki/One_of_These_Things_(Is_Not_Like_the_Others) Some of the actual games The idea is that there are four different objects, but one in particular stands out from ...
1 vote
1 answer
48 views

How can a Machine Learning model predict this classification problem?

I am pretty new to ML and have a basic but fundamental question. Let’s imagine we want to create a simple Sentiment Analysis model using Machine Learning not Deep Learning algorithms, so we need to ...
0 votes
2 answers
111 views

If Large Language Models like ChatGPT grow, do also their problems grow?

In a video Noam Chomsky said, that if these LLM get bigger, than also the things they are not good at get bigger too. He doesn't explain more details about this. So is this true? In what way do their ...
1 vote
1 answer
64 views

Automated theorem provers using models

I am wondering if they are automated theorem provers which can take as input known models of the theory, for example to help discard statements which are not correct. For example, if the statements ...
0 votes
0 answers
24 views

Facial Recognition

I need an algorithm that can do facial recognition in a video based on a training set. For any faces detected, it should give an output as how much the face matches with the faces in the training set. ...
1 vote
0 answers
79 views

Draw the decision boundary of an neural network

I'm really confused when it comes to what the difference between bias and threshold is. I have read that they are basically the same thing and that they serve the same purpose. Being on opposite sides ...
1 vote
0 answers
79 views

New Knapsack Problem solution with neural network approach

While I was learning the Constraint Satisfaction Problems, I analyzed the Knapsack problem(https://www.csplib.org/Problems/prob133/) and I saw the approach with the neural network was not the best ...

1
2 3 4 5
12