8

AlphaFold (version 1 and 2) predicts (so estimates) the 3D shape of the protein from the sequence of amino acids. AlphaFold's performance is measured with the global distance test (GDT), which is a measure of similarity between two protein structures (the prediction and the ground-truth) that ranges from 0 to 100. There is a short video and a longer one (...


6

A simple sanity-check on whether an image classifier can perform a task in theory is: Can a human expert, using the same image plus a list of catgeories that they are familiar with, perform the same task? It is important you only consider the contents of the image (or in general the data you are prepared to supply to the classifier) and the expert's ...


5

Why are CNNs useful? The main property of CNNs that make them more suitable than FFNNs to solve tasks where the inputs are images is that they perform convolutions (or cross-correlations). Convolution The convolution is an operation (more precisely, a linear operator) that takes two functions $f$ and $h$ and produces another function $g$. It's often denoted ...


5

They are all related terms. From top to bottom: One-shot learning aims to achieve results with one or very few examples. Imagine an image classification task. You may show an apple and a knife to a human and no further examples are needed to continue classifying. That would be the ideal outcome, but for algorithms. In order to achieve one-shot learning (...


5

In general, calculation of distance between camera and object is impossible if you don't have further scene dependent information. To my knowledge you have 3 options: Stereo Vision If you have 2 cameras looking at the same scene from a different point of view you can calculate the distance with classical Computer Vision algorithms. This is called stereo ...


5

The 2015 article Cyclical Learning Rates for Training Neural Networks by Leslie N. Smith gives some good suggestions for finding an ideal range for the learning rate. The paper's primary focus is the benefit of using a learning rate schedule that varies learning rate cyclically between some lower and upper bound, instead of trying to choose a single fixed ...


4

The update form $\theta^{\prime} \leftarrow \tau \theta+(1-\tau) \theta^{\prime}$ (where $\theta'$ and $\theta$ represent the weights of the target network and the current network, respectively) does exist and is correct. It is called soft update and it has been used in the Deep Deterministic Policy Gradient (DDPG) paper, which uses the concept of a target ...


4

Most model-fitting is stochastic, so you get different parameters every time you train, and you usually can't say that one algorithm will always give you a better-performing model. However, since you can retrain many times to get a distribution of models, you can use a statistical test like the T-Test to say "algorithm A usually produces a better model ...


4

I'll cover both L2 regularized loss, as well as Mean-Squared Error (MSE): MSE: L2 loss is continuously-differentiable across any domain, unlike L1 loss. This makes training more stable and allows for gradient-based optimization, as opposed to combinatorial optimization. Using L2 loss (without any regularization) corresponds to the Ordinary Least Squares ...


4

It seems your question is concerned with how an empirical mean works. It is indeed true that, if all $x^{(i)}$ are independent identically distributed realisations of a random variable $X$, then $\lim_{n \rightarrow \infty} \frac{1}{n}\sum_{i=1}^n f(x^{(i)}) = \mathbb{E}[f(X)]$. This is a standard result in statistics known as the law of large numbers.


4

I guess the issue is you lost track of where the samples came from and since you requested a math explanation I'll try to go step by step using my notation and without checking other material to avoid being biased by how other authors present it So we start from $$ L(D,G) = E_{x \sim p_{r}(x)} \log(D(x)) + E_{x \sim p_{g}(x)}\log(1 - D(x)) $$ then you apply ...


4

The visualisation can be found in The need for small learning rates on large problems. This paper by D. Randall Wilson and Tony R. Martinez from 2001 investigates the role of learning rates in gradient descent algorithms. In general, different algorithms assign different meaning to the same word 'learning rate'. For example, the learning rate in a gradient ...


4

Yes, it is not unusual to omit the bias by adding a neuron which always outputs a constant 1, which will then be multiplied by an appropriate weight to give the same formula as you would get using an explicit bias. One notable text using this convention is Understanding Machine Learning: From Theory to Algorithms by Shai Shalev-Shwartz and Shai Ben-David. ...


3

***Take my answer as a side note to that given by cantordust: If one can verify that an activation function perform well in some cases, that good behavior often extrapolates to other problems. Thus, by testing activation functions on a few different problems, one can often infer how well (or badly) it will perform on most problems. The following video shows ...


3

A neural network can be reduced to a linear regression model only if we use linear activation functions (i.e. $\sigma(x) = x$), and only if we do not use any neural network specific techniques such as convolution, residuals, etc., as shown below: $\text{neural network}(x) = \sigma_n(W_{n} \sigma_{n-1}(W_{n-1}\dots\sigma_1(W_1 x + b_1) + \dots + b_{n-1}) + ...


3

Let me answer your questions one by one. Submit it to a conference Let's start with the optimistic case. Say your paper gets accepted! You can upload your preprint on arXiv with the "arXiv.org perpetual, non-exclusive license to distribute this article (Minimal rights required by arXiv.org)". It is a non-Creative Common License that does not provide any ...


3

I know that a seed can be set to incorporate more determinism into the training. However, there could be other pseudo-random sequences that produce slightly better results? That is correct. If you fix the seed for a process which inherently has stochastic behaviour by design (such as initialising neural network params), then what you know about the model is ...


3

GPUs are able to execute a huge amount of similar and simple instructions (floating point operations like addition and multiplication) in parallel. In contrast to a CPU which is able to execute a few complex tasks sequentially very quick. Therefore GPUs are very good at doing vector & matrix operations. If you look at the operations performed inside a ...


3

Yes it should be possible. You may have a bug in your code, or the wrong hyperparameters. Training ResNet-50 will take a long time. Try training on other sets of images and see what accuracy you get to check if your approach is correct. Or, try loading a pretrained model, and training from that.


3

There are tasks in computer vision where recurrent neural networks (RNNs) can be useful because there's some sequential sub-task in the main task. For instance, in the paper Long-Term Recurrent Convolutional Networks for Visual Recognition and Description, the authors investigate the use of a neural network that is both recurrent and convolutional to solve ...


3

The paper that appears to have introduced the term "softmax" is Training Stochastic Model Recognition Algorithms as Networks can Lead to Maximum Mutual Information Estimation of Parameters (1989, NIPS) by John S. Bridle. As a side note, the softmax function (with base $b = e^{-\beta}$) $$\sigma (\mathbf {z} )_{i}={\frac {e^{-\beta z_{i}}}{\sum _{j=...


3

I don't think people generally do use neural nets for grid world. As long as the state and action spaces are small enough, you should be able to store Q values in a table like you suggested. Neural nets come in handy when the state space is very large (or even continuous), so you can't afford to store a table of Q values. Also, neural nets have the ability ...


3

As you stated, it's popular to have some form of a rectified linear unit (ReLU) activation in hidden layers and the output layer is often a softmax or sigmoid (depending also on the problem: multi-class or binary classification, respectively), which provides an output that can be viewed as a probability distribution. You could generalize this further to ...


3

In general, it is definitely very computationally expensive, so an exhaustive search is not performed in practice. However, there are some recent approaches for determining whether the architecture is "fine" without training the neural network first - by looking at the covariance matrix after forwarding the data, for example, in a recent paper ...


3

I'll answer in a couple of stages. I feel somewhat lost as to what the input for the NN should look like. Your choices boil down to two options, each with their own multitude of variants: Vector Representation: Your input is a vector of the same size as your vocabulary where the elements represent the tokens in the input example. The most basic version of ...


3

Depends on perspective. On one hand, you have an agent playing in an environment with another agent also evolving. This falls under the definition of Multi-Agent Learning, as can be seen with works such as Michael Bowling and Manuela Veloso. Multiagent learning using a variable learning rate. Artificial Intelligence, 136(2):215 – 250, 2002. Michael Bowling....


3

Deep learning is used to perform language translation in Google Translate [1]. Specifically, Google Translate now uses transformers and RNNs rather than the original GNMT system (proposed in 2016), which was also based on neural networks. Deep learning is also used in DeepL (though I cannot find a good resource to cite apart from Wikipedia [2] given that the ...


3

In a nutshell : Memorizing is not Learning So, first let's just remind the classical use of a neural net, in Supervised Learning : You have a set of $(x_{train}, y_{train}) \in X \times Y$ pairs, and you want to extract a general mapping law from $X$ to $Y$ You use a neural net function $f_{\theta} : x \rightarrow f_{\theta}(x)$, with $\theta$ the weights (...


3

In the article Playing Atari with Deep Reinforcement Learning, Mnih et al, 2013, which was a major outbreak in Deep Reinforcement learning (especially in Deep Q learning), they don't feed only the last image to the network. They stack the 4 last images : For the experiments in this paper, the function φ from algorithm 1 applies this preprocessing to the ...


3

The state space is certainly continuous, assuming that you can somehow feed that AI exact coordinates. You may have to resort to CNNs if you do not have access to this information. For the action space, you should consider how the game actually plays. Since you use a mouse to simply show the direction, you could use (x,y) positions of the mouse as an action, ...


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