# Tag Info

## Hot answers tagged neural-networks

15

Genetic Algorithms and Neural Networks are not suitable here. They are meta-heuristics for finding a good-enough, approximate solution to a problem. Notably, both require you to find a cost function to rate candidate solutions. Once you have such a cost function, it might be easier to manually come up with an algorithm that optimizes for this cost. This is ...

12

You can use simulated annealing. I did something like that before I landed my first job - see https://vimeo.com/20610875 (demo starting at 2:50, algorithm explained from 6:15). Simulated annealing is a type of a genetic algorithm, and maybe it was not suitable in theory (as @amon maintains in his answer), but it worked very well in practice, and it was ...

7

Work in binary: 0, 1, 10, 11, 100, 101... Know your math: 0+0=00, 0+1=01, 1+0=01, 1+1=10 Know your logic: or, and, not, xor... Find that the low bit is a XOR and the high bit is a AND. Expand the principle of one bit to 8, 16, 32, 64 bits Build it with logic gates. If you want to know more, see my answer to How is fundamental mathematics efficiently ...

7

If the data is linearly separable then yes, it's possible. Take one of these scatter plots which show the blue points and the red points and the line between them. (image stolen from here) If your neural network got the line right, it is possible it can have a 100% accuracy. Remember that a neuron's output (before it goes through an activation function) ...

6

Genetics Algorithms do apply here. During my undergraduate program, one of my colleagues wrote a paper to very similar problem of yours. You can look for Job Shop Scheduling and also Open Shop Scheduling or Flow Shop Scheduling can be interesting starting points To use a genetic algorithm you don't need a perfect solution, you can start with N random ...

6

Honestly, I don't think true TDD is a good fit for heavily stochastic programs. Really honestly, I don't think it's a good idea for much of anything, but putting that aside, you're going to make life harder than it needs to be trying to do GP where you have to fail tests before you allow yourself to write any code. There are a few ways to do good unit ...

6

You are correct about AI which includes ML which includes DL. NN can indeed be included in ML, be it inside or outside of the DL context. An example for the latter is when neuronal nets are used in simple task based learning (e.g. recognize car number plates in pictures). Data mining is somewhat broader than your definition, because it's not only ...

5

I'm not an neural network expert but I understand that identity mapping ensures that the output of some multilayer neural net is ensured to be equal to its input. Such a net is also called a replicator. I have understood that such identity/replication facilitates unsupervised training, and that the hidden layers of such nets can be used for feature ...

4

The crucial part of the story is this: Currently, I am in the process of building a crowd sourced platform for people who are knowledgeable to go in and mark up compatibility between those parts as its not always clear cut if they are for example: And this... Now what I want is an AI to be able to learn from the decisions of the crowdsourced ...

4

It's the same as it is in Algebra. An identity map or identity function gives out exactly what it got. When they say: h(xl) = xl They mean h is an identity mapping / function. If you give it xl it will give you back xl. h might be something else but once they say it's h(xl) = xl then it's an identity map / function. I don't see anything here to ...

4

Treat commas, periods and other forms of interpunctuation as if they were words. That allows you to train the neural network to learn when it is appropriate to end a sentence with a period or insert a comma. "End of text" should also be a "word" the neural network should learn to use appropriately. Turning a statement into a question is a task which is ...

3

The first approach is called supervised learning, and the second is called reinforcement learning. There are two ways you can use a neural network with reinforcement learning for chess: as a policy network or as a value network: a policy network would decide which move to play, whereas a value network would just evaluate the utility of a board position and ...

3

The practical uses of a neural network is pretty much everything. Recognition / detection in vision Artificial intelligence in games Classification ... In short : neural network can do pretty much everything as long you're able to get enough data and some efficient machine to get the right parameters. My professor told me once that some competition between ...

3

The fundamental thing you're doing is changing a weight from the wrong value to the right value. The fundamental thing that makes that happen is training. How much training is needed depends on how far from right your weight is and how much each training session changes that. Now sure, other factors can cloud that but if you're not doing measurements and ...

2

Well, I haven't built too many AI systems before, but let's take a naive crack at this. If I were going to use a neural network for this, the assigned tags to new parts would be the input nodes, the (vast) list of items it could be compatible with would the the output nodes, and the hidden layer would be it's user-confirmed compatibilities or ...

2

NNs are often used for tasks where we're unsure about what "features" are important. I can recognize a handwritten "2" but it's difficult to describe the essence of a "2" given the enormous variation in hand writing. In your case, the important features of your items seem to be decided by the tags. Humans have already done the hard work. Similarly NNs are ...

2

Have a Look in ada boost (adaptative booster). The idea is to use independent weak classifier (your detectors) and combine them in a clever way. It roughly goes like this: It first tries to evaluate the different detectors to detect and use an optimal linear combination of them to classify the data. It then tries to find another combination able to provide ...

2

It sounds like this would be suitable for a neural network, probably a standard feedforward type. I'm not sure how much you know about neural networks, but FYI the 'rules' it discovers won't be in a human-readable format. So if you want to run images through it and sort them, it'll do that; but if you're aiming to get a list of rules that you can see, you'...

2

"Hidden" nodes aren't really hidden like a black box - it is just a layer in between your input and output nodes. You program will have the values of all weights and the signal values propagated through the neural network. Once you have the output error, you can use all that information with the "archaic" (do you mean arcane?) mathematics. I found this ...

2

Question 1: Is my above understanding of the process correct? Yes; for the feed-forward process. It should be noted that the process can be computed with a matrix multiply for each layer. Question 2: Is this correct? Should the output values be a direct result of the sigmoid function? Yes, you generally apply the activation function to output layer. It ...

2

Well, someone upvoted this question and I guess that made it pop up for me. It's nice to know I learned something in 4 years :) Turns out the answer to this question is, in many wonderful and general ways, affirmative. Decoupled Neural Interfaces make NNs layer-wise asynchronous, allowing multiple concurrent forward and backward passes through a network, ...

2

What you are looking for is Neuro Evolution. According to this paper you could be in a position to create a trainable feed forward Neural Network (some extra information might also be found here). My recommendation would be to start small, maybe first start by making your character move and stop in the presence of danger. Neural networks can be rather ...

2

The symbolic methods of machine learning encompass both supervised and unsupervised learning. Supervised symbolic learning covers mining logical rules and dependencies from data: "if-then" rules decision trees and also learning concepts from data: mining ontologies from data hypothesis generation ... For unsupervised symbolic learning the well known ...

2

Randomizing the nodes and connections, increases the initial entropy of the network. Creating stronger biases in the network at the beginning (before training), would presumably facilitate the formation of some of the neural paths faster. When all other factors are equal for an input, this will amortize worst cases during training (this should happen ...

2

I found out some of the factors that may contribute to the effect. 1) At least in tiny-cnn, some of the buffers are allocated not once but once per worker thread. On a machine with 8 CPU threads, this can increase the memory usage a lot. In debug mode using MS VC++ 2015 the following two lines in the code base allocate a big chunk, both related to worker ...

2

You seem to be attempting to train your second layer's single perceptron to produce an XOR of its inputs. This isn't possible; a single perceptron can only learn to classify inputs that are linearly separable. The usual solution to solving the XOR problem with perceptrons is to use a two-layer network with the back propagation algorithm, so that the hidden ...

2

This is a very interesting project that you undertake. But it will also be very challenging Multi-agent architecture You have started to design a multi-agent system, with two active agents: the strategic agent and the tactical agent. Both feed the game control API on their own with commands. There is a "feedback loop" where agents send an input to ...

2

In general most sound processing works like other natural language processing in that one of the first steps is to slice your data into basic tokens, i.e. words - in human sound processing we split the words based on the silence between them. Accordingly you can pre-process to: Filter out sound outside of then normal, significant, speech bandwidth, this is ...

2

It's possible so long as you've done nothing to entrench the training. A neural network starts out weighed randomly. As you train it the weights gradually take on values that produce your desired output (through a feedback loop). Once you're satisfied you stop training. Now you want to add new things for the net to recognize. Fine. Either start randomly ...

1

Your calculation for the amount of memory used appears to be related to the number of neurons in the network and storing a double for each, but that isn't the only storage that is required -- each neuron will also contain a number of weights, each of which is likely to need at least a float. This is the last column in your output, and (at least if I ...

Only top voted, non community-wiki answers of a minimum length are eligible