# Tag Info

3

These are big areas, so here is a brief description of the differences: Game theory is concerned with studying solutions for 'games', which are basically a set of decisions leading to certain outcomes. In game theory you look at strategies to achieve the best outcome for a given participant. One classic example (which isn't really a game in the traditional ...

0

Not sure about where you can find datasets by difficulty, but I will concentrate on how you can generate your own spiral dataset. To generate a spiral you just need to create a time vector $t$ (e.g., a column in excel with numbers from 0 to $T_{max}$) and then use the following formulas (for the next two columns): x(t) = (r_0+v_rt)\cos(\omega t),\hspace{... 1 Not necessarily. Supposing your data is from the distribution of possible images containing an upright person close to the camera. Something like a neural network would perform poorly on the new data since it comes from a distribution other than on what it was trained. You could try augmenting the dataset to try to get some synthetic "far away upside down ... 3 The 'vocabulary' used is: \mathscr{F} - fluents, conditions that change over time. The predicate holds(<fluent>,<state>) defines whether a condition applies in a given state. \mathscr{A} - actions, describe, erm, actions that happen. The predicate occurs(<action>,<state>) specifies that a given action happens at a particular ... 3 There have been many methods proposed for text generating, but recurrent network dominates natural language processing with a key component: the perception of time. Many networks have been tried for text generation, with notable examples such as Markov chain. However RNN have been proven to work the best and is dominating the field of language modelling (... 2 Recurrent Neural Networks (RNNs) have been applied to generate text. In this blog post you will find a couple of interesting text examples (the author also has made his code available on github), e.g. their Shakespeare-like texts generated by an RNN: PANDARUS: Alas, I think he shall be come approached and the day When little srain would be attain'd ... 3 This course is focused on machine learning using MATLAB, which is not practical nowadays as it is a programming language used specifically for computing, and cannot display GUI or communicate through the network. The language is powerful but limited in some ways. Nowadays most people use python for machine learning, as it is versatile and can connect to ... 0 I must simply direct you to this excellent blog post on Machine Learning Mastery: https://machinelearningmastery.com/what-is-holding-you-back-from-your-machine-learning-goals/ 2 The most distinct words of a language are usually the function words (the, and, of, with,...); other lexical items are often (at least partly) shared between languages that had come in contact with each other. So looking for function words is usually the best way to identify the language in a given text. This can be done by having a list of function words ... 4 Calculation of gradient \begin{align} \nabla_{\theta} \log(\pi_{\theta}(a|s)) &= \phi(s,a) - \mathbb E[\phi (s, \cdot)]\\ &= \phi(s,a) - \sum_{a'} \pi(a'|s) \phi(s,a') \end{align} is only valid for linear function approximation with action preferences of form $$h(s, a, \theta) = \theta^T \phi(s,a)$$ and softmax policy \... 2 This could be possible, providing you have the right dataset to train it on. The volume of a cup consist of width, height and depth. You can probably detect all three of those given the bounding box or the pixels of the cup. However detecting the dimensions of an object require a reference object, like a penny or your finger and you have to specify the ... 0 very interesting questions: 1) what exactly is happening when training and validation accuracy change during training The accuracy change after every batch computation. You have 588 batches, so loss will be computed after each one of these batches (let's say each batch have 8 images). However, the accuracy you see in the progress bar it is the accuracy of ... 1 Hello and welcome to the community. There are multiple ways you can train a neural network: stochastic, mini-batch and batch. What you explained is the stochastic mode, where you input one training example 01 for example, calculate the gradients and update the networks weights before the next training example is fed. You could also select multiple such ... 2 I think you're looking for the minimization of false positives, that is, the instances that are classified as belonging to the desired class (the positive part of false positives) but that do not actually belong to that class (the false part of false positives). In practice, given your constraints, you may want to maximize the precision, while maintaining a ... 0 No pre-requisites required for Andrew Ng ML course. There are a couple of lectures in which he gives basic idea of Linear algebra. Also you can learn math when required. 1 The telltale signature of overfitting is when your validation loss starts increasing, while your training loss continues decreasing, i.e.: (Image adapted from Wikipedia entry on overfitting) It is clear that this does not happen in your diagram, hence your model does not overfit. A difference between a training and a validation score by itself does not ... 1 There are two main things to consider for dealing with imbalanced data: During Training: Undersampling the majority class (healthy patients) so that the model is not that biased to predicting healthy During Evaluation: Using a suitable metric to try to evaluate your model and try to optimize on when you are fine-tuning your random-forest. For imbalanced ... 1 You do not necessarily need to understand the concept of a random variable (r.v.) to understand the concept of a probability distribution, but the concept of a random variable is strictly connected to the concept of a probability distribution (given that each random variable has an associated probability distribution), so, before proceeding, you should get ... 2 A probability distribution in ML is the same as a probability distribution elsewhere. A probability distribution (or probability function, or probability mass function, or probability density function) is any function that accepts as input elements of some specific set x \in X, and produces as output, real-valued numbers between 0 and 1 (inclusive), such ... 3 The validation loss settles exactly at an error of one. Probably means there's something off with either the kind of data validation set has or with something in the training. An exact validation loss of one almost definitely means there's something off. I'd recommend before doing anything thoroughly go through your data or see if there's anything to debug ... 1 You are right CNN based models can outperform RNN. You can take a look at this paper where they compared different RNN models with TCN (temporal convolutional networks) on different sequence modeling tasks. Even though there are no big differences in terms of results there are some nice properties that CNN based models offers such as: parallelism, stable ... 0 I am working also on a similar issue. I think these readings could help. https://www.springer.com/gp/book/9783642004827 https://arxiv.org/abs/1211.0906 I think a problem that I faced is to elaborate some features of the problem, which has not been yet elaborated. Best regards. 3 Depends on what does 1 represent in your task. If you are trying to predict household prices and 1 represents \1, I think the average validation loss is good. If 1 represents \10000 in this case, probably something is not right. But remember that there are 2 parts contributing to the overall loss. The mse loss and the l2 penalty loss. (Also remember that ... 0 The best way is probably to Google it with "[org name] tensorflow github" and look what you get. For instance I found: Microsoft Nvidia Intel 2 Of course. It only depends if those features are informative enough for the task at hand. In order to better understand the phenomenon, you can imagine 2 features displayed as points in a 2D plane. The number of possible target classes goes up to the number of clusters you can find in that plane. About the suggestion, I can only recommend the utilisation of ... 0 In computational learning theory, a learning algorithm (or learner) A is an algorithm that chooses a hypothesis (which is a function) h: \mathcal{X} \rightarrow \mathcal{Y}, where \mathcal{X} is the input space and \mathcal{Y} is the target space, from the hypothesis space H. For example, consider the task of image classification (e.g. MNIST). ... 1 A random forest is a collection of classification trees. If more than 50% of these trees predict class A (and not class B), the random forest will predict class A. What you can do is lower the percentage needed to classify it as class A (in your case, patient has the virus). This way, you can tell your random forest to predict class A if only 20% (or 10%, ... -1 Just going to add this as another form of AI "learning". Genetic algorithms are different from neural networks. During my degree we had a focus on GAs so I'll explain our outlook on "teaching" a GA. It's less learning or teaching and more tuning. In the instance of our final year project in my AI class we were given 3 equations that needed solving, in each ... 1 You should only look for the cross-validation score. If this set is large enough, it will give you an accurate prediction of how your model will act for unseen data. Your case is exceptional. The fitted model which is obviously overfitted actually performs better on the cross-validation set. This means in turn that your overfitted model will perform better ... -1 Your question is exactly the definition of Unsupervised Learning. Unsupervised learning has many different methods to use, each of which is suitable for different types of problems. Therefore, read about unsupervised learning to find out which of them is appropriate for solving your problem. 1 What you are missing is what the news story does't mention and gloss over. When a news article says: company A has a large human face database so that it can train its facial recognition program more efficiently What it really means is: company A has a large database of human faces along with additional information such as the identity of the person ... 3 I think you're probably looking at this the wrong way around. A conventional, old-fashioned AI doesn't make a guess, then require confirmation as to whether that guess was right or wrong. Instead, (in the simplest case) it undergoes a one-off computationally intensive "training"/"learning" phase, during which you feed it an enormous number of correct answers ... 3 how can an AI be trained if we human beings are not telling it its calculation is correct? What you are looking for is called self-supervised learning. Yann LeCun, one of the originators behind modern neural network systems, has suggested that machines can reason usefully even in the absence of human-provided labels simply by learning auxiliary tasks, the ... -2 Also, the original database with images already has lots of info. Every image is linked to gender, name, age, and the fact that image is in fact a face. There is a possibility that the database in question has multiple images of the same person. At which point all answers are already there, all you have to do is to pose meaningful questions 1 A robust ML model is one that captures patterns that generalize well in the face of the kinds of small changes that humans expect to see in the real world. A robust model is one that generalizes well from a training set to a test or validation set, but the term also gets used to refer to models that generalize well to, e.g. changes in the lighting of a ... 2 Consider a dataset S \in \mathbb{R}^{N \times (M + 1)} with N observations (or examples), where each observation S_i \in \mathbb{R}^{M + 1} is composed of M elements, one value for each of the M features (or independent variables), f_1, \dots f_M, and the corresponding target value t_i. A decision tree algorithm (DTA), such as the ID3 ... 0 I can't remember the researcher's name, but he specializes in psychology in Great Britain and has done a lot of work with machine learning and artificial intelligence. The project he was working on that I read about earlier this year was one where they tried to deduce how humans learn. They came up with the theory that we learn by making guesses about ... 1 Both of them using the end-to-end approach for speech recognition. However, because of the code complexity in DeepSpeech, you can't tune the model for your work. Kaldi could be configured in a different manner and you have access to the details of the models and indeed it is a modular tool. I think Kaldi could be a better tool academically and also ... 0 The trick with unsupervised learning is that the AI doesn't learn that something is a face or not, it just sees unnamed patterns that the researchers need to then name. Let's say you feed it a dataset with one million pictures in order to train a facial recognition algorithm. After training, the AI will have found a few patterns in the pictures based on the ... 1 A hypothesis is a statement that suggests an as yet unproven explanation of a relationship between two or more phenomena that you intend to test. An agronomist thinks that more nitrogen on canola will always increase the crop outputHarvest = f(N)$$, or a meteorologist thinks he can show that the path of a hurricane over the ocean can be determined by ... 1 This is called decomposition of multi-class classifier. Your proposed method is called one vs all. One vs. all provides a way to leverage binary classification. Given a classification problem with N possible solutions, a one-vs.-all solution consists of N separate binary classifiers—one binary classifier for each possible outcome. During training, the ... 4 Taking your example of the faces data, keep in mind that when the model is run on a new unseen image the model can only return the already seen identity which emerges as the closest match. The result may be incorrect. The chances of mis-identification are much lower as the number of features incorporated increases. The input of the engineers lies at the ... 31 By "company A has a large human face database so that it can train its facial recognition program more efficiently" the article probably means that there is a training dataset S of the form$$ S = \{ (\mathbf{x}_1, y_1), \dots,(\mathbf{x}_N, y_N) \}  where $\mathbf{x}_i$ is an image of the face of the $i$th human and $y_i$ (which is often called a ...

2

I try to answer the things I know for sure: One effect of bigger images is the increasing computation time due to more pixels (input to your training) 4.Grayscaling reduces the information, which might decrease training time, but also model performance (accuracy, precision, recall). What I have seen is that grayscaling is used in for example face detection ...

2

Decision Tree learners, on their own, are not a good way to deal with imbalanced data. The most commonly used algorithms, by default, make no attempt to address this problem. If you look carefully at the three sources you post, you will find that they actually all agree on this point. Two of the sources actually propose methods of addressing this ...

1

You need to have access to the 696th hour (or successive hours), otherwise, you cannot test your model. An alternative would be, for example, to train your model on the first 693 hours, validate it on the 694th hour, and test it on the 695th hour.

2

From your description, the problem you want to solve is a linear optimization problem: suppose we use the indices $i$ and $j$ to denote the $i-$th class and the $j-$th grade. Also, let us call $y_j$ the current value of the $j-th$ grade, $g_j$ the goal value in grade $j$, $c_{ij}$ the impact factor of taking class $i$ in grade $j$, and $x_i$ the binary ...

1

In general, you can use a simulation to prepare and train a controller for a real world application. A good example of this being done for robotics is in the paper Autonomous helicopter flight via reinforcement learning where a Reinforcement Learning agent was trained on a model of helicopter dynamics before being used in reality. Often, as in this case, ...

4

Trying to address all the questions asked in the end in the same order Most definitely possible. I would say its best you approach this with segmentation to start with. Just use a free GPU runtime notebook service such as Google Colab or Kaggle Kernels. But you would not directly be able to integrate with the device, you'd have to keep moving input and ...

0

Some ideas out the top of my head: In the case of $dy/dx_2>0$ you could compute the gradient using the chain rule and limit the weights so that the constrain holds In the case of $y + x_5 + x_7 < K$ you could use a clipping function on the output layer?

Top 50 recent answers are included