72
votes
Accepted
Are neural networks prone to catastrophic forgetting?
Yes, indeed, neural networks are very prone to catastrophic forgetting (or interference). Currently, this problem is often ignored because neural networks are mainly trained offline (sometimes called ...
- 41.4k
23
votes
Are neural networks prone to catastrophic forgetting?
Yes, the problem of forgetting older training examples is a characteristic of Neural Networks. I wouldn't call it a "flaw" though because it helps them be more adaptive and allows for interesting ...
- 3,213
14
votes
Accepted
Is it possible to train a neural network as new classes are given?
I'd like to add to what's been said already that your question touches upon an important notion in machine learning called transfer learning. In practice, very few people train an entire convolutional ...
- 2,017
7
votes
What do you call a machine learning system that keeps on learning?
There are several terms or expressions related to such systems, such as
online learning
incremental learning
continuous learning
continual learning
lifelong learning
They are sometimes used ...
- 41.4k
7
votes
Is it possible to train a neural network as new classes are given?
Here is one way you could do that.
After training your network, you can save its weights to disk. This allows you to load this weights when new data becomes available and continue training pretty ...
- 850
7
votes
What is the difference between learning without forgetting and transfer learning?
Learning without Forgetting (LwF) is an incremental learning (sometimes also called continual or lifelong learning) technique for neural networks, which is a machine learning technique that attempts ...
- 41.4k
6
votes
Are neural networks prone to catastrophic forgetting?
What you are describing sounds like it could be a deliberate case of fine-tuning.
There is a fundamental assumption that makes minibatch gradient descent work for learning problems: It is assumed ...
- 191
6
votes
Can I train a neural network incrementally given new daily data?
Yes, this is possible. Continuously extending your training data is known as incremental learning.
You might also want to take a look at transfer learning, in which you reuse a trained model for a ...
- 176
4
votes
Are there dynamic neural networks?
This article on Dynamically Expandable Neural Networks (DEN) (by Harshvardhan Gupta) is based on this paper Lifelong Learning with Dynamically Expandable Networks (by Jeongtae Lee, Jaehong Yoon, Eunho ...
- 151
4
votes
Can a CNN be trained incrementally?
You are looking for incremental (or online) learning.
A CNN can be trained incrementally. For example, in the paper Incremental Learning of Convolutional Neural Networks, the authors propose an ...
- 41.4k
4
votes
Accepted
What are the state-of-the-art approaches for continual learning with neural networks?
What I understand from your questions is that you are trying to avoid catastrophic forgetting while applying online learning.
This problem should be addressed by implementing methods that reduce ...
- 253
3
votes
What are the state-of-the-art approaches for continual learning with neural networks?
Do you know which are the state-of-the-art approaches on this topic, and could you point me to some literature on them?
This answer already mentions some of the approaches. More concretely, currently,...
- 41.4k
3
votes
Accepted
Is continuous learning possible with a deep convolutional neural network, without changing its topology?
In general, is continuous learning possible with a deep convolutional neural network, without changing its topology?
Your intuition that it is possible to perform incremental (aka continual, ...
- 41.4k
3
votes
Is it possible to train a neural network as new classes are given?
There are several ways to add new classes to the trained model, which require just training for the new classes.
Incremental training (GitHub)
continuously learn a stream of data (GitHub)
online ...
3
votes
Are there dynamic neural networks?
I mostly studied HMMs and such models are called Infinite HMMs in that specific domain.
I believe that what you are looking for is called Infinite Neural Networks. Not having access to scientific ...
- 260
2
votes
Accepted
Will training an AI still work if the input data is somewhat sparse?
First, the title mentions "sparse data". Recently the expression has taken a clear meaning: The agent input is data with mostly zeros. In the question a different meaning: A "sparse data stream", ...
- 1,510
2
votes
What is the name of an AI system that learns by trial and error?
Near solution to your problem definition is reinforcement learning. You can define some reward using the objective function and define some possible state space for the machine and finally solve the ...
- 1,836
2
votes
What is the difference between learning without forgetting and transfer learning?
What I want to achieve is incremental training. So, as soon as I get new data, I can further train my already trained model and I don't have to retrain everything.
Learning without forgetting is one ...
- 2,819
2
votes
Are neural networks prone to catastrophic forgetting?
Maybe in theory, but not in practice. The thing is you seem to consider only chronological/sequential training.
And there are two ways to view this issue:
online learning -> then it is a feature of ...
- 121
2
votes
Accepted
How does a neural network that has been trained keep learning while in a real world scenario
You are right. If you don't continuously train the neural network after you have deployed it, there is no way it can continuously learn or be updated with more information. You need to program the ...
- 41.4k
2
votes
What are the state-of-the-art approaches for continual learning with neural networks?
There are lots of different approaches that try to avoid catastrophic forgetting in neural networks. It is impossible to summarize all contributions here.
However, in addition to the already mentioned ...
- 21
2
votes
How will MLOps and lifelong learning be complementary?
Lifelong learning and MLOps are indeed complementary.
Lifelong learning (LL) can be defined as the set of learning algorithms and models that can deal with more and more data and/or tasks without ...
- 41.4k
1
vote
Accepted
How to improve a trained model over time (i.e. with more predictions)?
That is exactly a neural network works like.
Suppose you have a 1000 examples. How you train a network is: First, you divide these 1000 into maybe 100 batches (10 each). After that's done, you feed a ...
- 868
1
vote
Is there any real-time computer vision system that can learn to detect new objects of new classes?
You are probably looking for incremental learning (sometimes known as lifelong learning) techniques, i.e. machine learning techniques that attempt to address the catastrophic forgetting effect of ...
- 41.4k
1
vote
Transfer learning to train only for a new class while not affecting the predictions of the other class
Even if you want to re-train your model for just one new class you will have to prepare your training data such that it includes all or most of the classes which you want to predict. Most of the times ...
- 111
1
vote
What are the most common methods to enable neural networks to adapt to changing environments?
For the vast majority of cases where you have a dynamic(and assumed non-linear) relationship between your input and output, you would not use modified architecture. You would simply retrain on the new ...
- 1,260
1
vote
How does FastText support online learning?
The pull request #1327 (https://github.com/facebookresearch/fastText/pull/1327)
Allows for:
test after each epoch
checkpointing
training on large data which does not fit into memory (largest I tested ...
1
vote
Accepted
What is the name of an AI system that learns by trial and error?
I believe "Reinforcement Learning" is the term you are looking for (as mentioned by others as well) but keep in mind that the scope of your problem falls under the section of AI that is called Search.
...
- 241
1
vote
What is the name of an AI system that learns by trial and error?
I think any learning algorithm probably uses trial and error and analysis of the results with the ultimate goal of maximizing utility.
It seems that the recent milestones in AI fall under the general ...
- 6,219
1
vote
Is it possible to train a neural network as new classes are given?
You could use transfer learning (i.e. use a pre-trained model, then change its last layer to accommodate the new classes, and re-train this slightly modified model, maybe with a lower learning rate) ...
- 41.4k
Only top scored, non community-wiki answers of a minimum length are eligible