Newest Questions
12,769 questions
2
votes
1
answer
49
views
Applications of Manifolds to Artificial Intelligence
In summary: what applications of smooth manifold theory exist to Artificial Intelligence?
I have some knowledge of smooth manifold theory, and was wondering how I could apply it to a project on ...
- 121
1
vote
0
answers
37
views
How to make a variational autoencoder work on time-freq matrix?
I want to use a complex valued variantional autoencoder for unsupervised blind source separation. As an input to the network, I am giving the time-freq matrix of the spectrogram instead of the ...
- 41
0
votes
0
answers
20
views
CLIPSeg: no change in performance metrics with a better convolutional decoder
I am training CLIPSeg on the Oxford IIIT pet dataset for semantic segmentation (3 classes: background, cat, dog). In short, what I do is I stick a decoder on the CLIP encoder. The encoder outputs:
...
- 1
1
vote
1
answer
34
views
Method to generate blindly/exhaustively
Much like image generators can be mined for unique and interesting outputs prompt free, is there a way to blindly generate high probability, unique texts? I’m trying to make a movie script generator ...
- 11
0
votes
1
answer
30
views
Can an encoder-only transformer integrate with an decoder-only one?
E.g. an encoder only model understands the text and a decoder only model provides details about it.
- 377
3
votes
2
answers
50
views
Required background for thorough understanding of Causal ML research papers?
I'm interested in pursuing research in the intersection of causal inference and machine learning, particularly on causal discovery and causal representation learning. Through my exploration so far, I ...
0
votes
0
answers
22
views
why's there Nan values for forecast and total loss?
So I am training a Graph attention based model on time series dataset(Swat) for which while evaluating the dataset function for it is
...
1
vote
1
answer
42
views
Should encoder only or decoder only transformers be used for name origin identification?
Like, identifying what mythology is a god from by providing it's name.
- 377
1
vote
0
answers
16
views
find leader voters based on their voting patterns
Suppose we want to identify leader voters (the most influential voters) based on their voting patterns across different subjects. What method would you recommend that also guards against ...
2
votes
1
answer
92
views
Learning curve behaviors across double descent regimes
I am learning about double descent phenomenon from here: https://www.di.ens.fr/~fbach/learning_theory_class/lecture9.pdf
I was asking myself:
When training a system, how can we know in which regime ...
- 255
0
votes
0
answers
24
views
What is the best method to teach a new syntax to an AI model using Vertex AI?
I would like to teach a new language to an AI model like Gemini. This language is used in a third party tool, syntax is a modified form of TCL with the tool's own commands. What is the best way to ...
0
votes
0
answers
112
views
Connect Open WebUI with Langflow using Open WebUI Function Pipe
I am trying to connect Langflow to Open WebUI so that I can use Langflow workflows inside Open WebUI as a model. This makes handling all the AI easy because its all based on Open WebUI.
Below is the ...
- 1
0
votes
0
answers
29
views
Is it normal to get zero error for a class in a binary classification problem using Genetic Programming?
I am trying to do highly unbalanced binary classification using Linear Genetic Programming to detect a certain spoken word. I use mel coefficients as features. The instructions include basic ...
5
votes
1
answer
83
views
Is PAC-unlearnability a fundamental limitation for LLM reasoning?
For simplicity, let’s focus on knowledge reasoning tasks with Yes/No answers. According to learning theory, even moderately complex knowledge reasoning tasks are PAC-unlearnable. This implies that no ...
- 180
4
votes
1
answer
78
views
Why Are the Standard and Markov Chain Derivations of the Policy Gradient Theorem Equivalent?
While studying the proof of the Policy Gradient Theorem, I have come across two different approaches.
The first seems to be a more standard approach involving "unrolling" across every time ...
