New answers tagged

2

In addition to the ways the term topology is itself used generically to describe the "shape" of various aspects of Machine Learning, the term appears in the field Topological Data Analysis: In applied mathematics, topological data analysis (TDA) is an approach to the analysis of datasets using techniques from topology. Extraction of information from ...


0

What are mono-variable and multi-variable neural networks? I am not sure about this, because most (if not all useful) neural networks are multivariable neural networks (i.e. they contain multiple parameters). Even the perceptron usually contains more than one parameter, so that terminology isn't clear even to me. Maybe they are referring to the number of ...


0

Adding to the current answers. Definitions of Artificial Intelligence can be categorized into **four categories Thinking Humanly, Thinking Rationally, Acting Humanly and Acting Rationally. The following picture (from Artificial Intelligence: A Modern Approach) will shed light on over these definitions: But what is more interesting is the AI effect. ...


0

The usage of the word "kernel" in the context of support vector machines probably comes from its usage in the context of integral transforms. See the article Kernel of an integral operator, and the questions What is the difference between a kernel and a function? and Why is the kernel of an integral transform called kernel?. The word "kernel" has been ...


0

Though the word "Advantage" in the actor-critic realm has been used to refer to the difference between the state value and the state action value, A2C brings in the ideas of A3C. In A3C, several worker networks interact with different copies of the environment (asynchronous learning) and update a master network after a set if steps. This was meant to solve ...


2

Well, one of the simpler definitions for SI sounds like this: The emergent collective intelligence of groups of simple agents.” (Bonabeau et al, 1999) So, in order to get to the SI you have to use some kind of algorithms/AI to get simple intelligent agents. It's just cooperative intelligence, or cooperative AI if you wish. SI just uses today's AI/ML ...


1

Held-out simply means "not included" particularly in the sense of: This part of the data was not included in this specific training run. Depending on the context of all of these text non-held-out data/classes means the data that actually was included in a particular modeling exercise. Consider this excerpt from your first example: For instance, Owen ...


2

In Supervised learning, the goal is to learn a mapping from points in a feature space to labels. So that for any new input data point, we are able to predict its label. whereas in Unsupervised learning data set is composed only of points in a feature space, i.e. there are no labels & here the goal is to learn some inner structure or organization in the ...


3

Supervised learning The supervised learning (SL) problem is formulated as follows. You are given a dataset $\mathcal{D} = \{(x_i, y_i)_{i=1}^N$, which is assumed to be drawn i.i.d. from an unknown joint probability distribution $p(x, y)$, where $x_i$ represents the $i$th input and $y_i$ is the corresponding label. You choose a loss function $\mathcal{L}: ...


2

Let's have a look at the introduction of Chapter 2: Multi-armed Bandits in the Reinforcement Learning: An Introduction by Sutton, Barto The most important feature distinguishing reinforcement learning from other types of learning is that it uses training information that evaluates the actions taken rather than instructs by giving correct actions. This ...


1

Several important researchers distinguish between bandit problems and the general reinforcement learning problem. The book Reinforcement learning: an introduction by Sutton and Barto describes bandit problems as a special case of the general RL problem. The first chapter of this part of the book describes solution methods for the special case of the ...


0

OK, I think I understood what this means. Hard and easy negatives are ones which have a relatively large and small values for the loss function respectively.


1

Not in terms of models, but there is a terminology called 'Hierarchical learning', wherein if your model has a task to classify disease, then, If it detects a presence of a disease (disease/ no disease), then it proceeds to further classify a disease(class A/B/C/...). Else it does not proceed. This technique of hierarchical learning is very common amongst ...


0

The words latent space and embedding space are often used interchangeably. However, latent space can more specifically refer to the sample space of a stochastic representation, whereas embedding space more often refers to the space of a deterministic representation. This comes from latent referring to an unobserved random variable, for which we can infer a ...


3

Lowest layer generally refers to the layer closest to the input. This comes from the idea that layers closer to the input represent low-level features such as gradients and edges, while layers closer to the output represent high-level features such as parts and objects.


2

People sometimes use 1st layer, 2nd layer to refer to a specific layer in a neural net. Is the layer immediately follows the input layer called 1st layer? The 1st layer should typically refer to the layer that comes after the input layer. Similarly, the 2nd layer should refer to the layer that comes after the 1st layer, and so on. However, note that this ...


3

The term feature embedding appears to be a synonym for feature extraction, feature learning etc. I.e. a form of embedding/dimension reduction (with the caveat the goal may not be a lower dimensional representation but one of equal dimensionality, but more meaningfully expressed): Feature embedding is an emerging research area which intends to transform ...


Top 50 recent answers are included