I haven't found resources from people working on object recognition of 3d cad models. I'll pose a problem I'd like to be able to solve: I have a library of cad models containing thousands of combustion engines. I need a program that can recognize where the crank shaft axis is in 3d space. I think a convolutional neural network could be programed to recognize 3d solids. By turning a cad model into a voxelated version of the part, you could slice the data and present the data similarly to 2d images. Though, I'm not sure how you could convolute a zero-thickness mathematical construct such as an a ..read more

I'm a passionate third-year Computer Engineering student, and I'm reaching out in pursuit of valuable advice. Lately, I've noticed that most of my peers have made remarkable strides in machine learning, natural language processing (NLP), and deep learning. Their expertise is truly inspiring, and it has ignited a strong desire within me to not only catch up but also excel in these fields. However, I must admit that there are times when I feel overwhelmed by the sheer complexity and vastness of machine learning and NLP. The rapid pace of advancement and the depth of knowledge required can be dau ..read more

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 allocating indivisible items (goods or bads) under capacity constraints. In this setting, we are given a set of categorized items. Each category has a capacity constraint (the same for all agents), that is an upper bound on the number of items an agent can receive from each category. Our main result is a polynomial-time algorithm that solves the problem for two agents with additive utilities over the i ..read more

The model I'm building first selects a secret word at random from a list. The model which uses an API then returns a row of underscores (space separated)—one for each letter in the secret word—and asks the user to guess a letter. If the user guesses a letter that is in the word, the word is redisplayed with all instances of that letter shown in the correct positions, along with any letters correctly guessed on previous turns. If the letter does not appear in the word, the user is charged with an incorrect guess. The user keeps guessing letters until either (1) the user has correctly guessed al ..read more

I had an idea where I want to train a model based on two other neural networks (nn). These two NN share the same objective goal, but have differing levels of performance, for ht objective task, as they are trained on different domains. Is there any precedent for a model where competitively train the two models against each other, and combine the best of both results adversarially? Is there precedent on this type of machine learning model ..read more

Is there a way to train huge models (1+ TiB of parameter data) being distributed on web? For example, I have a several hundreds of budget machines, each equipped with 2-4 GiB of RAM and with no powerful GPU. Is there a ready developed software that supports this scenario? If such a software doesn't exist could you point to some relevant researches ..read more

I have scrapped Glassdoor Reviews. The reviews are split into two columns: Pros : What the employees liked about the company Cons : what the employees didn't like about the company After all the necessary text pre-processing from removing stopwords, punctuation, symbols, performing stemming and lemmatization and visualizing wordclouds , I thought about performing Topic Modeling algorithms for more comprehension of my data , since wordclouds are not enough. For this , according to my small knowledge , I should be testing multiple algorithms. I chose LDA, NMF , HDP & LSA. Now to find out the ..read more

Does anyone use a platform for ML Observability for Recommender Systems and can tell me what are the main advantages of this? And if you don't use it then why did you choose not to ..read more

I would like to find some sort of Chatbot model that I can layer my own 'x' training data on top of to make it sort of like an 'x' centric chatbot. The idea would be to be able to add as much 'x' training data to it as possible, on top of the existing chat oriented data. I'm hoping somebody knows of something like this ..read more

In the picture below there are some regions which are very bright (i.e. more white). Some bright regions are wide and some are narrow or thin. The red box covers one such wide bright spot, and blue box covers one thin bright spot. Thin bright spots are called edges and wide bright spots are called hot-spots. I want to remove all the hot-spots from the image (i.e. make them black), but no edge should be removed. My question is how to write Python code using OpenCV to remove all hot-spots but no edge ..read more