Questions tagged [performance]
For question about methods of evaluating the performance (e.g. the accuracy) of an artificial intelligence algorithm or model.
57
questions
2
votes
1
answer
1k
views
Should I be layer freezing when fine-tuning an LLM?
I've had it in my head that generally speaking, it's better to freeze layers when fine-tuning an LLM, as per this quote from HuggingFace's article:
PEFT approaches only fine-tune a small number of (...
0
votes
0
answers
81
views
Typical FLOPs for training and hyperparameter tuning of standard models?
I am trying to find estimates of the typical FLOPs required for training a modern model. The FLOPs used for inference are widely reported and cluster around $10^6$ - $10^7$ for many large models. ...
- 119
1
vote
1
answer
131
views
What does that mean if my precision, F1-score are very high, but my ROC AUC score is around 0.5?
What does it mean when my precision and so on are so high and the roc auc score is around 0.5?
0
votes
2
answers
190
views
How do you interpret this train vs test accuracy scores? is the model under or over fitting?
What does this difference in train and test accuracy mean?
0
votes
1
answer
81
views
Classifier performance if data are deterministic
so let us imagine one has a classification problem at hand, say objects with $n$ numeric features, to be classified as belonging to two classes ${0,1}$.
Data could look like, for $n=3$
...
- 45
0
votes
2
answers
57
views
Is the accuracy the best metrics to evaluate the performance of Deep Learning model? [closed]
Consider a model A that achieved an test accuracy of 99% on dataset-A with the size of 200 images and a model B that achieved ...
1
vote
1
answer
524
views
Why does mean episode reward during training differ dramatically from "manual" runs of the trained model on same data?
I am training an RL agent, using PPO, on a time-series environment that comes from a tabular dataset. The possible scores during an episode goes from -1 to positive infinity (though realistically, I ...
- 342
2
votes
1
answer
273
views
What is a 'degenerate run' in evaluating model performance?
I've recently come across a paper that uses the term "degenerate run", but I'm not sure if I understand what it means. The idea is that when they report the average performance of running ...
- 121
0
votes
0
answers
157
views
Neural network for an output in the form of a probability distribution
I am not an expert in machine learning. Recently, I want to construct a data-driven model based the neural network. The problem is that I want from my algorithm to learn an output in the form of a ...
0
votes
1
answer
129
views
Confusion Matrix Measures vs Accuracy level in Neural Network Model
I'm working on a classification machine learning problem with two classes: high and low, which are derived from another numerical column x. Previously, if x>100, the sample is considered ...
- 241
2
votes
1
answer
465
views
Why doesn't the high precision of neural network weights improve accuracy?
Consider the following paragraph from the subsubsection 3.5.2: A dtype for every occasion chapter named It starts with a tensor from the textbook titled Deep Learning with PyTorch by Eli Stevens et al....
- 3,612
0
votes
0
answers
277
views
GAN performance starts to get worse as training continues
I'm currently trying to train a GAN to recreate similar images from a dataset. The dataset is using the Eiffel Tower Pictures from Googles Quick Draw dataset. The images aren't very large (only 12x12 ...
1
vote
1
answer
594
views
Test accuracy decreases during my train process
I want to train a neural network model with the arcface loss function and try to combine it with domain adaption. But when the training process continues, I find the test accuracy first increases and ...
- 11
0
votes
1
answer
114
views
Do I need to tune the hyper-parameters or more data if SVR model performs poorly?
I am using non-linear data to SVR and have tried tuning the hyperparameters and still have a poor model performance. Do I need more data or format the data for more suitable results?
I get similar ...
1
vote
1
answer
141
views
What could cause the hamming loss and subset accuracy to get stuck in a multi-label image classification problem?
I am rather new to deep learning and got some questions on performing a multi-label image classification task with keras convolutional neural networks. Those are mainly referring to evaluating keras ...
- 11
1
vote
0
answers
107
views
Are goal-reaching and optimizing the utility function special cases of performance measure?
In AIMA, performance measure is defined as something evaluating the behavior of the agent in an environment.
Rational agents are defined as agents acting so as to maximize the expected value of the ...
- 205
0
votes
1
answer
43
views
Will there be any changes in the model's performance due to the usage of very small batch sizes?
I am trying to run a code that has a batch size around 28. I can run the program on my GPU with this batch size.
But, when I modify the code for my requirements and try to run, it is showing an run-...
- 3,612
0
votes
1
answer
3k
views
How do I know what a good mean absolute error value is? [closed]
I have just run an MAE calculation for my machine learning models and the results show:
SVM MAE = 28.850 deg.
Random Forest MAE = 33.832 deg.
How do I know what a good MAE value is? What is the ...
0
votes
0
answers
97
views
Identifying if a model is over or under-fitting via graphs
I am working on a Neural Network and have plotted the performance of my model. However the plots seem not to fit the "trends" (which help you identify the issue with your model) presented in ...
2
votes
1
answer
383
views
How would the performance of federated learning compare to the performance of centralized machine learning when the data is i.i.d.?
How would the performance of federated learning (FL) compare to the performance of centralized machine learning (ML), when the data is independent and identically distributed (i.i.d.)?
Moreover, what ...
- 35
1
vote
0
answers
65
views
Is the performance of a neural network, which was trained with encrypted data and weights, affected if the weights are decrypted?
Suppose that a neural network is trained with encrypted (for example, with homomorphic encryption and, more precisely, with the Paillier partial scheme) data. Moreover, suppose that it is also trained ...
- 73
1
vote
1
answer
150
views
Can the attention mechanism improve the performance in the case of short sequences?
I am aware that the attention mechanism can be used to deal with long sequences, where problems related to gradient vanishing and, more generally, representing effectively the whole sequence arise.
...
- 11
3
votes
1
answer
200
views
How should I change the hyper-parameters of the C51 algorithm, in order to obtain higher reward?
I have a scenario where, in an ideal situation, the greedy approach is the best, but when non-idealities are introduced which can be learned, DQN starts doing better. So, after checking what DQN ...
- 185
0
votes
1
answer
742
views
How is the performance of a model affected by adding a ReLU to fully connected layers?
How significant is adding a ReLU to fully connected (FC) layers? Is it necessary, or how is the performance of a model affected by adding ReLU to FC layers?
4
votes
2
answers
239
views
How do you measure multi-label classification accuracy?
Multi-label assignment is the task in machine learning to assign to each input value a set of categories from a fixed vocabulary where the categories need not be statistically independent, so ...
- 251
1
vote
0
answers
25
views
Would performance of atomic models matter in ensemble methods?
Suppose I have two fitted ensemble models $F_1 := (f_1, f_2, f_3, \cdots f_n)$ and $G_1 := (g_1, g_2, g_3, \cdots g_n)$.
And they were using the same ensemble methods (boosting or bagging).
And I am ...
- 111
2
votes
0
answers
58
views
What is human-level performance for semantic segmentation?
I see so many papers claim to have an algorithm that beats 'human-level performance' for semantic segmentation tasks, but I can't find any papers reporting on what the human-level performance actually ...
1
vote
0
answers
34
views
How is the performance of a CNN trained with monochrome images on image recognition tasks degraded?
For CNN image recognition tasks, like object recognition/face recognition/object segmentation/posture recognition, are there experiment results about how much will the performance be degraded with ...
- 199
0
votes
1
answer
951
views
Which kind of data does sigmoid kernel performance well?
While I was playing with some hyperparameters, I came to a wired situation. My dataset is IRIS dataset to be specific. SVM algorithm has some hyperparameters that we can tune, such as Kernels, and C ...
1
vote
0
answers
65
views
How many training runs are needed to obtain a credible value for performance?
I'm trying to optimize a neural network. For that, I'm changing parameters like the batch size, learning rate, weight initialization, etc.
A neural network is not a deterministic algorithm, so, in ...
- 131
2
votes
4
answers
9k
views
Is it normal to have the root mean squared error greater on the test dataset than on the training dataset?
I am new to deep learning.
I am training a model and I am getting a root mean squared error (RMSE) greater on the test dataset than on the training dataset.
What could be the reason behind this? Is ...
- 69
1
vote
1
answer
97
views
How to estimate the accuracy upper limit of any CNN model over a computer vision classification task
We are given a computer vision classification task, that is, a task that asks us to predict the category of an image over $n$ predefined classes (the so-called closed set classification problem).
...
- 111
2
votes
0
answers
39
views
Can the addition of low-quality images to the training dataset increase the network performance?
I already trained a deep neural network called YOLO (You Only Look Once) with high-quality images (1920 by 1080 pixels) for a detection task. The result for mAP and IOU were 93% and 89% respectively.
...
- 21
1
vote
1
answer
2k
views
How can I merge outputs of two separate layers so that the overall performance improves?
I am training a combined model (fine-tuned VGG16 for images and shallow FCN for numerical data) to do a binary classification. However, the overall AUC score is not what I expected it to be.
Image-...
- 241
1
vote
0
answers
40
views
Sample size for the evaluation of Deep Learning Models
I'm evaluating the performance and accuracy in detecting objects for my data set using three deep learning algorithms. In total there are 24,085 images. I measure the performance in terms of time ...
0
votes
0
answers
27
views
What are the properties of a model that is well suited for for high performance real-time inference
What are general best practices or considerations in designing a model that is optimized for real-time inference?
3
votes
2
answers
2k
views
Why is the effective branching factor used for measuring performance of a heuristic function?
For search algorithms with heuristic functions, the performance of heuristic functions are measured by the effective branching factor ${b^*}$, which involves the total number of nodes expanded ${N}$ ...
- 165
1
vote
1
answer
63
views
What will change when workstations will have ARM Machine Learning Processors onboard?
lately we read that many manufacturers are forcing ARM architectures to be used on future workstations. One of ARM's recent announcements is a machine learning processor. What will change in terms of ...
- 141
5
votes
1
answer
78
views
How can we conclude that an optimization algorithm is better than another one
When we test a new optimization algorithm, what the process that we need to do?For example, do we need to run the algorithm several times, and pick a best performance,i.e., in terms of accuracy, f1 ...
- 51
1
vote
0
answers
31
views
Improving Recall of a Certain Class
Let's say that we have a test data set with $20,000$ observations for which we want to make a binary prediction for. When we apply our best trained model to this data set (e.g. logistic regression ...
- 21
4
votes
1
answer
579
views
How to evaluate an RL algorithm when used in a game?
I'm planning to create a web-based RL board game, and I wondered how I would evaluate the performance of the RL agent. How would I be able to say, "Version X performed better than version Y, as ...
- 613
2
votes
0
answers
26
views
In addition to matrix algebra, can GPU's also handle the various Kernel functions for Neural Networks?
I've read a number of articles on how GPUs can speed up matrix algebra calculations, but I'm wondering how calculations are performed when one uses various kernel functions in a neural network.
If ...
- 131
1
vote
0
answers
162
views
Why does precision-recall curve become more stable when neural net begins to overfit?
I am training a convLSTM with a dropout layer (with prob 0.5).
If I train over more than 5 epochs I notice that the network starts to overfit: my validation set loss becomes stationary while the ...
- 130
5
votes
2
answers
187
views
How do I improve accuracy and know when to stop training?
I am training a modified VGG-16 to classify crowd density (empty, low, moderate, high). 2 dropout layers were added at the end on the network each one after one of the last 2 FC layers.
network ...
- 125
3
votes
1
answer
252
views
What are the aspects that most impact on the inference time for neural networks in embedded systems?
I work with neural networks for real-time image processing on embedded softwares and I tested different architectures (Googlenet, Mobilenet, Resnet, custom networks...) and different hardware ...
- 269
8
votes
1
answer
740
views
Why isn't the ElliotSig activation function widely used?
The Softsign (a.k.a. ElliotSig) activation function is really simple:
$$ f(x) = \frac{x}{1+|x|} $$
It is bounded $[-1,1]$, has a first derivative, it is monotonic, and it is computationally ...
- 183
1
vote
0
answers
70
views
Performance measure on windowed time series data
I have time series data where I use a sliding window to detect anomalies in those windows. A sliding window is an interval of the dataset that steps one datapoint for each iteration. Datapoints are ...
- 11
2
votes
1
answer
155
views
How can we compare the intelligence of AI systems?
One way of ranking human intelligence is based on IQ tests. But how can we compare the intelligence of AI systems?
For example, is there a test that tells me that a spam filter system is more ...
- 816
1
vote
0
answers
39
views
How to visualize/interpret text prediction model results?
I am using LSTM model to predict the next xml markup from an input seed.
I have trained my model on 1500 xml files. Each xml file is generated randomly. I am wondering if there is a way to visualize ...
- 169
1
vote
1
answer
141
views
How do I check that the combination of these models is good?
I've selected more than 10 discriminative (classification) models, each wrapped with a BaggingClassifier object, optimized with a ...
- 177