Questions tagged [optimization]
For questions about implementing and improving optimization algorithms used in creating AI programs, or optimization in general.
194
questions
0
votes
1
answer
57
views
How does MAML inner loop optimization works?
I started to learn meta-learning, reading the MAML paper https://arxiv.org/pdf/1703.03400.pdf
In the inner loop, I am calculating adapted parameters for each task, I will be doing multiple steps of ...
CommunityBot
- 1
1
vote
0
answers
31
views
Do NNs suffer from lack of efficiency in network structure and suggesting training parameters?
I am working on dynamical systems using Optimal Control theory and trying to find the connection between this field and Machine Learning. Consider a simple 2-layer Neural Network (NN) where the ...
0
votes
0
answers
24
views
Are there more optimization methods like GAE for PPO [closed]
I posted about this earlier, but got the suggestion to separate the questions.
I'm currently trying to "solve" the OpenAI gym "Humanoid" environment. To improve the training ...
- 33.8k
0
votes
0
answers
46
views
What method to use when optimizing an array of data
Say I have an array of data, where each element describes a shape made of points, in vector form (each vector has several hundred dimensions). Each element also has a rating that gets higher, the ...
1
vote
1
answer
53
views
Does reaching the global optima guarantee good performance in a task?
It is to my understanding that, in deep learning, we are essentially trying to minimize the loss function that we have defined and reach its global optima through some form of optimization technique. ...
- 451
2
votes
0
answers
23
views
How to create a loss function that penalizes duplicate indices in the output tensor?
We're working on a sequence-to-sequence problem using pytorch, and are using cross-entropy to calculate the loss when comparing the output sequence to the target sequence. This works fine and ...
- 121
5
votes
2
answers
2k
views
What are the advantages of the Kullback-Leibler over the MSE/RMSE?
I've recently encountered different articles that are recommending to use the KL divergence instead of the MSE/RMSE (as the loss function), when trying to learn a probability distribution, but none of ...
- 402
3
votes
3
answers
1k
views
In logistic regression, why is the binary cross-entropy loss function convex?
I am studying logistic regression for binary classification.
The loss function used is cross-entropy. For a given input $x$, if our model outputs $\hat{y}$ instead of $y$, the loss is given by
$$\text{...
1
vote
3
answers
140
views
How to improve neural network training against a large data set of points with varying magnitude
I am currently using TensorFlow and have simply been trying to train a neural network directly against a large continuous data set, e.g. $y = [0.014, 1.545, 10.232, 0.948, ...]$ corresponding to ...
CommunityBot
- 1
0
votes
0
answers
25
views
How to generate data lying in the union of different hyperplanes using a VAE
I know that a way to possibly encode prior knowledge into neural networks training is by using differentiable optimization layers (paper).
I'm in the following situation, and I'm wondering if it could ...
- 197
0
votes
0
answers
27
views
Resolving Derivation Discrepancies for Differentiating through Optimization Paths
I'm reading the paper "Optimizing Millions of Hyperparameters by Implicit Differentiation". The key contribution of the paper is to show that you can replace optimizing through the ...
- 1
1
vote
1
answer
60
views
Why is Adam trapped in bad/suspicious local optima after the first few updates?
In the paper On the Variance of the Adaptive Learning Rate and Beyond, in section 2, the authors write
To further analyze this phenomenon, we visualize the histogram of the absolute value of ...
CommunityBot
- 1
2
votes
1
answer
93
views
Are Genetic Algorithms suitable for a problem with a non-unique optimal solution?
I was wondering if a genetic algorithm is useful if the optimization problem has several optimal solutions.
My thought was that I should not use it since when combining two members of a population who ...
CommunityBot
- 1
1
vote
0
answers
28
views
Structured policies in dynamic programming: solving a toy example
I am trying to solve a dynamic programming toy example. Here is the prompt: imagine you arrive in a new city for $N$ days and every night need to pick a restaurant to get dinner at. The qualities of ...
- 11
4
votes
1
answer
49
views
How to use MOPSO to align characters vertically?
I need to efficiently align characters vertically using Multi Objective PSO. Alignment is achieved by adding spaces in between a given set of characters.
...
CommunityBot
- 1
5
votes
1
answer
4k
views
How can we use linear programming to solve an MDP?
Apparently, we can solve an MDP (that is, we can find the optimal policy for a given MDP) using a linear programming formulation. What's the basic idea behind this approach? I think you should start ...
- 33.8k
4
votes
1
answer
597
views
When training a CNN, what are the hyperparameters to tune first?
I am training a convolutional neural network for object detection. Apart from the learning rate, what are the other hyperparameters that I should tune? And in what order of importance? Besides, I read ...
CommunityBot
- 1
0
votes
2
answers
105
views
How can I model this problem of delivering assets by choosing a route with reinforcement learning?
I would like to build a model based on reinforcement learning (RL) for the following scenario
Recommend the best route (of cities listed for a given country) that satisfies the required criteria (...
CommunityBot
- 1
0
votes
0
answers
18
views
What is the use of utilizing q-Gaussian mutation operators in evolutionary algorithms?
There are great numbers of evolutionary algorithms for optimization of engineering problems which each of them gives its own objective function value in a defined problem. Using the q-Gaussian ...
- 23
1
vote
0
answers
43
views
Best algorithm for the Word Ladder puzzle
What would be the best performing algorithm to solve the Word Ladder problem, in terms of guaranteed finding of the shortest solution in the shortest possible time? Is it BFS, DFS, A*, IDA* or another ...
- 11
0
votes
1
answer
151
views
How to make input variable as trainable parameter in a neural network?
I am working on an optimization problem. First, I have done forward training to work the network as a surrogate model, then I freeze the output and I want to find an optimal value of input for a given ...
CommunityBot
- 1
1
vote
1
answer
203
views
How to properly optimize shared network between actor and critic?
I'm building an actor-critic reinforcment learning algorithm to solve environments. I want to use a single encoder to find representation of my environment.
When I share the encoder with the actor ...
CommunityBot
- 1
2
votes
1
answer
172
views
Continuous state and continuous action Markov decision process time complexity estimate: backward induction VS policy gradient method (RL)
Model Description: Model based(assume known of the entire model) Markov decision process.
Time($t$): Finite horizon discrete time with discounting factor
State($x_t$): Continuous multi-dimensional ...
CommunityBot
- 1
0
votes
0
answers
32
views
Using neural network to find the most similar items
I am trying to design a multilayer neural network that based on item features will return x amount of the most similar items from the dataset. My current neural network takes features of 2 items, and ...
- 1
0
votes
0
answers
17
views
How can TRPO with constrained form allow larger update step?
There are two optimization forms of TRPO.
One is that:
\begin{equation}\max\limits_{\theta}[L_{\theta_{old}}(\theta) - CD^{max}_{KL}(\theta_{old}, \theta)]\end{equation} where $C = \frac{4\epsilon\...
3
votes
1
answer
67
views
Why does my regression-NN completely fail to predict some points?
I would like to train a NN in order to approximate an unknown function $y = f(x_1,x_2)$. I have a lot of measurements $y = [y_1,\dots,y_K]$ (with K that could be in the range of 10-100 thousands) ...
- 31
1
vote
0
answers
47
views
Are Bayesian Optimization Methods Better Suited Noisy Optimization Problems?
We know that in many applied contexts (e.g. Machine Learning, Loss Functions for Neural Networks), the functions we are trying to optimize are "noisy" by definition (unlike in the classical ...
- 247
1
vote
1
answer
25
views
Additional Optimizations for Convolutional Models On Inferencing
I am aware of several ways to optimize a convolutional (or any) model after training to make inferencing quicker. I am currently implementing BatchNormalization Folding and removing Dropout layers ...
- 3,503
-2
votes
2
answers
62
views
What does it mean by Generalization? [closed]
Towards Theoretically Understanding Why SGD Generalizes Better Than ADAM in Deep Learning
What does it mean by Generalization in this article?
- 134
0
votes
0
answers
41
views
What all does the gradient tells us other than the direction to move parameters?
Gradients are used in optimization algorithms.
I know that a gradient gives us information about the direction in which one needs to update the weights of a neural network. We need to travel in the ...
- 3,099
0
votes
0
answers
22
views
Does Stochastic Gradient Descent "Work" on (Some) Non-Convex Functions?
As we know, there has been a lot of work and research done to demonstrate that the Gradient Descent Algorithm can converge on (deterministic) convex, differentiable and Lipschitz Continuous functions :...
- 247
0
votes
1
answer
34
views
Are Problems in AI Usually "Ill Posed"?
I was reading the following link (https://en.wikipedia.org/wiki/Well-posed_problem) on "Well Posed Problems". Supposedly, if a problem is "Well Posed", it must meet the following ...
- 739
0
votes
1
answer
192
views
Effects of ReLU Activation on Convexity of Loss Functions
I have heard the following argument being made regarding Neural Networks:
A Neural Network is a composition of several Activation Functions
Sigmoid Activation Functions are Non-Convex Functions
The ...
- 3,503
0
votes
0
answers
30
views
Can we compute probability of sample in R1-Natural Evolution Strategy with linear time and space?
In R1-NES, the sample is drawn from a multivariate normal distribution, $\vec{\theta} \sim \mathcal{N}\left( \vec{\mu},\mathbf{\Sigma} \right)$, with the covariance matrix is parametrized by $s$ and a ...
- 143
0
votes
0
answers
17
views
Can we use Stochastic Gradient Descent and its derivatives for geographical researches?
I am a freshman AI engineering student actually working on a project where I have to analyze a big amount of geographical data to find out an optimal position of latitude and altitude for a specific ...
- 33.8k
0
votes
0
answers
26
views
How to calculate the gradient (or derivative) of y = f(x) of y w.r.t x where y represents the order statistics divided by median of x?
How to calculate the gradient (or derivative) of y = f(x) of y w.r.t x where y represents the order statistics divided by median of x?
For instance x is ...
- 101
1
vote
1
answer
437
views
What is uncentered variance and how it becomes equal to mean square in Adam?
I have been reading about Adam and AdamW (Here). The author mentioned that in "uncentered variance" we don't consider subtracting mean
In this statement, the author is talking about ...
- 364
1
vote
0
answers
148
views
Why does the schema theorem of genetic algorithms hold?
I have been reading about the Schema Theorem - one of the first theorems from the field of evolutionary computing and genetic algorithms, largely responsible for justifying the use of genetic ...
- 33.8k
0
votes
0
answers
25
views
What I Should Do to Reduce Solution Size for Simulated Annealing Algorithm?
I am trying to find the best solution for radar placement problem with using multi objective simulated annealing algorithm.
So there is an area (in real map) and I want to put minimum count of radar ...
1
vote
1
answer
95
views
In the update rule of RMSprop, do we divide by a matrix?
I've been trying to understand RMSprop for a long time, but there's something that keeps eluding me.
Here is a screenshot from this video by Andrew Ng.
From the element-wise comment, from what I ...
- 33.8k
0
votes
0
answers
31
views
Limit of momentum update equation
I am self-studying on optimization algorithm on https://d2l.ai/chapter_optimization/momentum.html and couldn't get my head around some derivation:
Instead of the standard gradient descent update ...
- 101
4
votes
1
answer
118
views
Can I compute the fitness of an agent based on a low number of runs of the game?
I'm developing an AI to play a card game with a genetic algorithm. Initially, I will evaluate it against a player that plays randomly, so there will naturally be a lot of variance in the results. I ...
- 33.8k
1
vote
1
answer
28
views
Cover a surface with smaller predefined objects
I'm trying to make a program that takes a surface designed by the user, and different 3D geometries from a dataset as inputs and gives a good approximation of the surface using only the objects found ...
CommunityBot
- 1
0
votes
0
answers
51
views
Is it possible to find a good neural network structure without training it? [duplicate]
Neural networks consist of so many parameters. Researchers could create as many possible neural networks as they wish. So I want to ask a general question. Could we devise an evolutionary algorithm ...
- 33.8k
1
vote
0
answers
24
views
a loss for binary step function data
I have some data with ground truth that looks like a binary step function, where part of it is 0 and part is one.
An example for the GT can be like ...
- 111
2
votes
1
answer
206
views
If Deep Learning is non convex, then why do use a convex loss function?
I was just reading through some convex optimization textbooks to hopefully improve my deep learning understanding and come up with new ideas. Halfway through, I decided to Google a bit! It's obvious ...
- 33.8k
0
votes
0
answers
34
views
Why is there a Hessian diagonal approximation? And when can we use it?
This topic has been introduced in "Pattern Recognition and Machine Learning, Bishop, 2006", section 5.4.1. I am a bit confused about this method and I have two questions.
Why this method ...
0
votes
1
answer
1k
views
What are the advantages and disadvantages of using LISP for constraint satisfaction in 3D space
We are currently working on developing a 3D modeling software that allows designers to set spatial constraints to models. The computer then should generate a 3D mesh conforming to these constraints.
...
CommunityBot
- 1
3
votes
2
answers
83
views
Is there a way to define the boundaries of the optimal size of a training set?
At a related question in Computer Science SE, a user told:
Neural networks typically require a large training set.
Is there a way to define the boundaries of the "optimal" size of a ...
- 33.8k
0
votes
0
answers
36
views
How to smooth a cost function?
I have a combinatorial optimization problem whose loss function is really unsmooth now.
Without specifying my problem in detail, I was wondering if there are some general methods/steps I can apply to ...
- 33.8k