Questions tagged [online-learning]
Online learning refers to courses, books, forums, tutorials, and videos on data science or machine learning topics/concepts available online.
56
questions
0
votes
0
answers
10
views
What kind of learning do I need ? (use-case specific)
Consider a scenario where I have a model trained on gesture videos (say a 3D ResNet). I am looking for a technique (or a combination) that allows me to further train the model every time I have a new ...
- 139
1
vote
1
answer
66
views
What are the advantages of model drift vs concept drift in online learning?
I have asked this question here but I'm also posting it here to get a better insight:
https://stats.stackexchange.com/questions/602282/what-are-the-advantages-of-model-drift-vs-concept-drift-in-online-...
- 130
1
vote
0
answers
16
views
Epochs for new batch when online training?
I am online training a RNN with fixed batch size k on a time series. Initially I train my model with n batches and a number of e epochs. When a new batch n+1 is available, I would like to update the ...
- 11
1
vote
1
answer
465
views
Online Learning/Continual Learning for tree-based Algorithms
Every example I come across any kind of iterative learning on Random Forest/XGBoost/LightGBM, it just continuously grows the number of estimators for new batches of data by ...
- 328
0
votes
1
answer
40
views
Which machine learning models allow online training and which don't?
I am working on a project where I have to update my model every time I get feedback x times. For example, showing an Advertisement on an App and then, when the person doesn't click on in it after ...
- 1
0
votes
1
answer
23
views
Are most deep learning models online learning models?
I'm online learning starter. from my perspective, online learning model is the model which can update its paramater with data flows(I've seen a article pointing out that incremental model is ...
- 1
1
vote
1
answer
105
views
Trouble understanding regression line learned by SGDRegressor
I am working on a demonstration notebook to better understand online (incremental) learning. I read in sklearn documentation that the number of regression models that support online learning via the <...
- 289
3
votes
2
answers
1k
views
Difference between regret and pseudo-regret definitions in multi-armed bandits
I posted this question Cross Validated, but didn't get any answer. So I am posting it here too, as the question is very relevant to machine learning
I am following the book Bandit Algorithms. In page ...
- 141
2
votes
1
answer
100
views
Resources on on-line machine learning
I am wondering if there are any books/articles/tutorials about "on-line machine learning"?
For example, this website has nice lecture notes (from lec16) on some of the aspects:
https://web....
- 127
0
votes
1
answer
24
views
If we train a model every time from scratch by using current task and samples from memory (ER) then is it correct way to perform continual learning?
Suppose that there are T tasks. We use an experience replay (ER) strategy using a tiny episodic memory. Here, we train a model always from scratch at each task using current task samples and samples ...
- 301
1
vote
1
answer
56
views
Can Online DQN model overfit?
I am new in the area of RL and currently trying to train an online DQN model. Can an online model overfit since its always learning? and how can I tell if that happens?
- 11
1
vote
0
answers
19
views
For Incremental Learning ML Model do we have to perform any kind of label encoding?
Please guide me on Online / Incremental Learning ML model, I am using Creme tool for my hands-on, where as my dataset has some categorical features, I did tried to do encoding but still getting error ...
- 111
1
vote
0
answers
134
views
The impact of Normalization/Data Transformations on Incremental Learning
Suppose we have a Neural Network (or any machine learning model), and the goal is to perform incremental learning as new data comes in on a regular schedule.
There is extensive literature showing ...
- 826
20
votes
4
answers
24k
views
Incremental Learning with sklearn: warm_start, partial_fit(), fit()
I have built an ML model with the goal of making predictions for targets of the following week. In general, new data will come in and be processed at the end of each week and be in the same data ...
- 826
2
votes
0
answers
208
views
Retaining past learning with Incremental Learning [closed]
I have built an ML model with the goal of making predictions for targets of the following week. In general, new data will come in and be processed at the end of each week and be in the same data ...
- 826
5
votes
1
answer
2k
views
What is the difference between continual learning and active learning?
As per my understanding, active learning is a kind of continual learning. Is there any difference between them?
- 301
0
votes
1
answer
293
views
Incremental modelling of kmeans in pyspark
I have a large dataset and trained the model with kmeans for the first time. I saved the model and pipeline used . Now again I started collecting data. After sufficient data is collected using old ...
2
votes
1
answer
2k
views
SGDClassifier partial_fit() for online learning - is one step of gradient descent enough?
I'm interested in incremental (online) learning for my logistic regression model trained with SGDClassifier. Basically updating the model as more labeled data comes ...
- 181
3
votes
1
answer
45
views
Understanding experiments in Continual Learning
Via paper Continual Learning Through Synaptic Intelligence, I see this figure for Split MNIST benchmark, but there is a point I can get.
Here there are 5 tasks, ...
- 235
2
votes
1
answer
361
views
Distinguish Multi-Task vs Single-incremental Task in Continual Learning
I read through the Internet and found this:
Most of Continuous Learning studies focus on a Multi-Task scenario, where the same model is required to learn incrementally a number of isolated tasks ...
- 235
1
vote
1
answer
165
views
In Incremental Learning will the model be updated automatically?
I came across Incremental Learning algorithms paper, where incremental algorithms are compared. I have problem with general understanding. Will the model be updated /adapts itself automatically when ...
- 111
10
votes
4
answers
2k
views
SGDClassifier: Online Learning/partial_fit with a previously unknown label
My training set contains about 50k entries with which I do an initial learning. On a weekly basis, ~ 5k entries are added; but the same amount "disappears" (as it is user data which has to be deleted ...
- 111
1
vote
0
answers
535
views
SGDClassifier accuracy MUCH worse than LinearSVC
with LinearSVC, I get an accuracy off 0.89:
...
- 111
1
vote
0
answers
42
views
What strategies and algorithms are suited for using the time wasted in collecting big data?
Most state of art algorithms right now is using/exploiting big data. My concern is what can you do to maximize reward while waiting for large amount of data that you think is appropriate.
For ...
- 106
2
votes
1
answer
344
views
Comparison between approaches for timeseries anomaly detection
After various days of research, I could take a global picture of the existing methods to perform anomaly detection on time series, namely:
Forecasting with Deep Learning. Eg. RADM or LSTM model
...
- 153
2
votes
1
answer
341
views
Forecasting vs non-forecasting predition for time series anomaly detection
I have got the objective of implementing a uni/multivariate online anomaly detection system.
After multiple days of research, I could collect many ways to achieve this (Eg. moving average solutions ...
- 153
5
votes
1
answer
1k
views
First order vs Second order online machine learning algorithms
In this git repo the online learning algorithms are classified as first order and second order. I tried searching what it means, yet I'm unable to understand the different between such first oder and ...
- 51
2
votes
1
answer
67
views
Work experience in Data Science: Couple hours with A-level student [closed]
Our company has accepted an A-level student (18-year old, just before uni) for some work experience. The student will be with me for a couple of hours. I do data analysis and I'm quite busy at the ...
- 29
0
votes
1
answer
93
views
Predicting crowd density using real time streaming data
I would like to make predictions about how crowded a location (postal code mapped to lat/lng coords) would be. I would like to provide predictions to questions like: what’s the crowd density going to ...
- 11
1
vote
1
answer
86
views
Should I update my regularisation L1 and L2 regularisation parameters in online setting?
I have been working on online learning for a few weeks now, especially with Vowpal Wabbit and logistic regression. My understanding of the online learning algorithms and the problem is alright but I ...
- 178
5
votes
2
answers
689
views
How to continue incremental learning when a categorical variable has been assigned additional category labels?
Please help answer this question or point me to any resource.
There is a model in an environment where training happens with new data and the data is discarded after training is completed.
This keeps ...
1
vote
1
answer
262
views
Online vs minibatch training for speed
If I do online learning in a setting where I have a HUGE amount of data, is that faster than doing minibatch learning (even if I optimize my batch size for GPU use, that is, use a multiple of 32 ...
- 1,981
1
vote
1
answer
217
views
Clarifying my understanding of on-policy RL (online SARSA)
I want to clarify I have understood how SARSA works in nuances. Consider an original definition taken from ON-LINE Q-LEARNING USING CONNECTIONIST SYSTEMS. G. A. Rummery & M. Niranjan. CUED/F-...
- 233
1
vote
1
answer
555
views
contextual bandits for online learning
Which of the algorithms in the current literature for contextual bandits can be implemented for online learning and which ones can't? I'd really appreciate it if someone could provide a link to papers ...
- 171
2
votes
1
answer
239
views
Reasoning for temporal difference update rule
In TD(0) learning where the value function is given by $V(s) = w^T\phi(s)$ where $w$ is a weight vector and $\phi$ is a feature map, the weight update is given by $w_{t+1} = w_{t} + \eta\delta_{t+1}\...
2
votes
1
answer
692
views
Testing Multi-Arm Bandits on Historical Data
Suppose I want to test a multi-arm bandit algorithm in the contextual setting on a set of historical data. For simplicity, let's assume there are only two arms A and B and suppose the rewards are ...
- 171
3
votes
3
answers
7k
views
How to retrain the neural network when new data comes in?
I am new to deep learning. Can anybody help me with the online learning implimentation for deep learning models. As per my understanding, i can save a keras/tensorflow model after training and when ...
0
votes
1
answer
189
views
Recommend tags for documents
I'm working on a unusual issue (for me) and I need some advice.
My goal is to have a recommendation algorithm that propose tags for a document, based on all the previously tagged documents. For ...
- 181
1
vote
1
answer
788
views
how to retrain model with periodic new features?
I've trained a gradient boosting classification model. But, suppose i've a set of fixed features F1,F2....Fn and new features which are added weekly (no. of actions done in that week). So, after 2 ...
- 157
1
vote
0
answers
69
views
How to calculate inverse of square matrix for streaming or online data as all data are not available at once?
Suppose initial data is $D$ and need to calculate the inverse of covariance of matrix $D$ i.e.
$C = cov(D,D)$, where $cov$ represents covariance.
$B = inv(C)$
Now, new data $N$ appears. So matrix D ...
- 301
4
votes
1
answer
9k
views
Real time noise removal using Savitzky-Golay Method
I would like to ask if Savitzky-Golay can be implemented on real-time data.
I have used it on a fixed array size, but would like to extend it to output values for real-time sensor data. Can anyone ...
- 161
1
vote
2
answers
136
views
Online learning w/ feature weighting/adjusting
Let's say I have a supervised learning problem with a sequence of features and labels. First, I learn on the training data and then I decide to stream in data, point by point and do online learning. ...
- 13
2
votes
0
answers
524
views
How do I train a contextual bandit policy?
Say I'm attempting to improve click-through rates on videos on my website. I've been reading the literature on contextual bandits and came across the Microsoft MWT white paper. I believe this is the ...
- 301
1
vote
0
answers
279
views
Online methods for sequence prediction
A ml beginner here, so please bear with me. If I understand correctly RNNs seem to be the go to method right now for sequence prediction for a given input (single/as a sequence). But I do not have ...
- 11
2
votes
1
answer
206
views
Finding a proper method for an online driving style classification using acceleration data
I am using a smartphone in my car to gather acceleration data (both longitudinal and lateral). Now, I want to classify my data in real-time based on the acceleration force applied through accelerating,...
- 251
5
votes
2
answers
3k
views
online detection of plateaus in time series
I need to detect plateaus in time series data online. The data I am working with represents the magnitude of acceleration of a tri-axis accelerometer. I want to find a reference time window that I can ...
- 251
1
vote
1
answer
469
views
Can FTRL be applied on linear least squares? or is it just for logistic regression models?
I'm exploring follow-the-regularized-leader FTRL proximal gradient descent:
paper, reference implementation.
Everywhere FTRL is mentioned, the loss surface for the gradient decent is the ...
- 1,357
7
votes
0
answers
512
views
differences between LSQR and FTRL when working with very sparse data
I have a 2M instances dataset with millions of very very sparse dummy variables created using the hashing trick = ...
- 1,357
5
votes
2
answers
286
views
is neural networks an online algorithm by nature?
I have been doing machine learning for a while, but bits and pieces come together even after some time of practicing.
In neural networks, you adjust the weights by doing one pass (forward pass), and ...
3
votes
1
answer
485
views
Online/incremental unsupervised dimensionality reduction for use with classification for event prediction
Consider the application:
We have a set of users and items.
Users can perform different action types (think browsing, clicking, upvoting etc.) on different items.
Users and items accumulate a "...
- 83