New answers tagged

1

Since $A_t$ is already determined (because we are calculating $Q(S_t,A_t)$), I think $\pi(A_t|S_t)$ is definitely 1. But what about $\mu (A_t|S_t)$? Is it 1 or not? You could assign values of 1 to each to get the right answer, but the situation is different. You can see that more clearly in the definition of action value, $q(s,a)$: $$q_{\pi}(s,a) = \mathbb{...


0

One can use simpler approach with deepC compiler and convert exported onnx model to c++. Check out simple example at deepC compiler sample test Compile onnx model for your target machine Checkout mnist.ir Step 1: Generate intermediate code % onnx2cpp mnist.onnx Step 2: Optimize and compile % g++ -O3 mnist.cpp -I ../../../include/ -isystem ../../../packages/...


0

Chiming in because I had the same question and stumbled across your post. It seems like the general version of your question still has not been answered. In general, a well-formed gradient update rule is all you need to be able to train the network. We are thinking of converting to a "loss function" because that is the typical flow in the structure ...


0

The wiki has a concise quote by Andreas Hein, where the gap is defined by "the difference in meaning between constructs formed within different representation systems". This connotes the core problem of translating meaning between an informal language (typically natural language) and a formal language (programing language or other formal symbolic ...


4

In terms of transfer learning, semantic gap means different meanings and purposes behind the same syntax between two or more domains. For example, suppose that we have a deep learning application to detect and label a sequence of actions/words $a_1, a_2, \ldots, a_n$ in a video/text as a "greeting" in a society A. However, this knowledge in Society ...


-2

To some extend, it get rid of low intensity numerical noise. Condition properties of the optimization problem is always an issue, i suspect BatchNorm alleviate this instability.


1

The required shape of the tensor $T$ depends on the shape of other tensors that are involved in the same operations of that same tensor $T$ and the required/desired shape of the resulting tensor, in the same way that the number of columns of the matrix $M \in \mathbb{R}^{n \times m}$ needs to match the number of rows of the matrix $M' \in \mathbb{R}^{n' \...


1

I commonly use softmax for all 2-class or k-class problems, basically, because I always like to have an output node for each class. For sigmoid, i.e., logistic, you cannot estimate MSE for each sample using the relationship $E_i = \sum_c^C (y_c - \hat{y}_c)^2$, where $C$ is the number of classes, $y_c$ is 0 or 1 for true class membership, and $\hat{y}_c$ is ...


1

Sigmoid is used for binary cases and softmax is its generalized version for multiple classes. But, essentially what they do is over exaggerate the distances between the various values. If you have values on a unit sphere, apply sigmoid or softmax on those values would lead to the points going to the poles of the sphere.


1

Accuracy can sometimes be a very coarse metric. When it is applied to three class problems, people often take the class label with maximum predicted probability and predict that. The probabilities of the individual labels are ignored. I'd recommend that as well as accuracy you calculate sensitivity and specificity for each class and the area under the ROC ...


1

This seems to be a known problem, and intuitively seems reasonable. You might be interested in the paper Adversarial Training Can Hurt Generalization. The authors suggest that this might be because training on the perturbed data requires the model to learn more robust features, which means more samples are required to obtain performance comparable to a model ...


0

If the auto-encoder is converging to the same encoding for different instances, there may be a problem in the loss function. Check the size and shape of the output of the loss function, as it may be getting confused and evaluating the wrong tensors (i.e. you may need to transpose something somewhere). Basically, assuming you are using an auto-encoder to ...


1

It looks like your network is overfitting, because the training loss carries on decreasing to zero even though validation loss levels off, and then starts to increase again. I would guess that your network is essentially "memorising" the training examples because you're getting a near zero loss in training. You could try: applying some form of ...


1

You are talking about model parallelism. But, that's not the reason RNNs/LSTMs are not in vogue. Imagine your ability to read the first line of a page and going on reading and still making connections to the first line until the end of the page. Can RNNs/LSTMs do that? No. Can Attention (i.e. Transformers) do it? Yes. The reason is simple Attention is ...


1

It's my understanding that selecting for small models, i.e. having a multi-objective function where you're optimizing for both model accuracy and simplicity, automatically takes care of the danger of overfitting the data. Sort of. A secondary objective function often works as a form of regularisation, and can work to reduce overfit. However, this ...


1

Testing machine learning programs is quite different than testing traditional software. The main reason why this is the case is quite simple, if you're familiar with machine learning. ML programs are not just if statements and loops, but they are composed of models, which can even be black-box models, such as neural networks (i.e. it's difficult to interpret ...


0

Here is how I understand this regularization. $R_1$ is simply the norm of the gradients, which indicates how fast the weights will be updated. Gradient regularization penalizes large changes in the output of some neural network layer. $$ R_{1}\left(\psi\right) = \frac{\gamma}{2}E_{p_{D}\left(x\right)}\left[||\nabla{D_{\psi}\left(x\right)}||^{2}\right]\text{,}...


0

Yes, you can. Let's say you have 5 classes named a,b,c,d,e. You fit your data into a SVM Classifier and a Random Forest Classifier. Assume that, SVM classified "a" and "b" class well and RFC classified "c","d","e" well. So, ensembling these two models is going to increase accuracy dramatically. Ensemble ...


0

To speak to your question about how Chinese to English translation can be a computation, it first requires a way to turn the base units of translation (tokens) into something computable. One basic way is to define the set of your vocabulary terms and create a gigantic matrix (typically called an embedding) with each column representing a token as well as one-...


2

A function is simply a procedure that maps a particular input to a particular output. You put in $X$, and the function computes $Y$. Those $X$ and $Y$ can take many different forms. It could be mapping one number to another number (convert miles to kilometres), mapping sound to text (name that tune), mapping text to text (translate languages), mapping a ...


1

One way to look at intelligence is it's the way to compress the universe. That means we have a short mental representation of meaningful concepts. For example, if I would say "there is a red swan in your building, it's dangerous and can kill you", you already have concepts of "red", "swan", "danger" and this easy ...


1

That equation is just an assumption that we make about the relationship between a response variable (aka dependent variable) $y$ and a predictor (aka independent variable) $x$, i.e. the response variable (target) is an unknown function $f$ of the predictor $x$ plus some noise $\epsilon$ due to e.g. measurement errors (caused e.g. by damaged sensors). So, if ...


3

Not necessarily. The neural network (or whatever else you use) is a model of what you are trying to do, and usually models are not able to perfectly model reality, as it is too complex. A noise term is generally used to represent that, ie the imperfection of the model's relationship with the actual world.


0

A simpler answer is that for a standard neural net, the asymptotic behaviour is the asymptotic behaviour of the output neurons. For example, if the output layer is ReLUs, then the asymptotic behaviour is necessarily linear. In your case, since you want it to be asymptotically constant, you can use the slightly old-fashioned choice of sigmoid units in the ...


1

For standard NNs, their extrapolation behavior an important aspect for financial applications cannot be controlled due to complex functional forms typically involved. Neural Networks with Asymptotics Control discuss how they overcome this significant limitation and develop a new type of neural networks that incorporate large-value asymptotics, when known, ...


0

RTS smoother can be divided into two parts. In first part, for data points $i = 0, 1, \ldots, N$ you do standard Kalman filtering and you get a sequence of estimates $\hat x_0, \hat x_1, \ldots, \hat x_N$ Second part of RTS smoother works backwards from $i = N-1, \ldots, 1, 0$ with \begin{equation} \hat x^s_N = \hat x_N \tag{1} \end{equation} thats because ...


1

We want a distribution over $w$, don't we? Yes. You want to obtain a distribution over the parameters, which models the uncertainty about the parameters. This distribution over the parameters can induce a probability distribution over the possible functions consistent with your data. Why is $a$ integrated out here and not $w$? This is just the definition ...


0

You have two dependent variables $a$ and $w$. So, there is a joint distribution $p(w, a)$. You can make a marginalization by one of them, pretty much as you did in your second formula. $$p(w) = \int p(w, a)da$$ $$p(w) = \int p(w | a)p(a)da$$ The only difference in this case, the calculation made for the specific point $x_i, y_i$, which is empathized by sub-...


1

There are some works that do this comparison. Briefly, it's been observed that the performance of models trained via FL drops as data distributions between participating agents differ. When data is IID-like though, performance is comparable to centralized training. Some works that I'm aware of are as follows: Overcoming Forgetting in Federated Learning on ...


Top 50 recent answers are included