- Transfer learning reuses a model trained on one task as the starting point for a related task.
- Transfer learning can cut training time from weeks to hours.
- In computer vision, early and middle layers that detect edges and shapes are reused, and only later layers are retrained.
- Types of transfer learning include domain adaptation, domain confusion, multitask learning, one-shot learning and zero-shot learning.
- Transfer learning is popular in computer vision and NLP, where accurate models need large amounts of data.
Creating complex models from scratch requires vast amounts of computing resources, data, and time. Transfer learning accelerates the process by leveraging commonalities between tasks (such as detecting edges in images) and applying those learning to a new task. Training time for a model can go from weeks to hours, making machine learning more commercially viable for many businesses.
Transfer learning is very popular in domains like computer vision and NLP where large amounts of data are needed to produce accurate models.
How does Transfer Learning work?
For example, in computer vision, neural networks usually detect edges in the earliest layers, shapes in the middle layers, and some task-specific features in the later layers. The early and middle layers are used in transfer learning, and only the latter layers are retrained. As a result, it helps leverage the labeled data of the task it was trained on.
What are the types of Transfer Learning?
Below are the five different types types of transfer learning:
1. Domain adaptation
The concept of domain adaptation is usually applied when the marginal probabilities between a source domain and a target domain are different.
2. Domain confusion
In a deep learning network, different layers capture various features. We can identify domain-invariant features and improve their transferability between domains. The model is then nudged to learn as similar a representation as possible.
3. Multitask learning
Multitask learning is a slightly different flavor of transfer learning. Multitask learning involves learning several tasks simultaneously without distinguishing between sources and targets. Unlike transferring learning, where the learner initially knows nothing about the target task, the learner receives information about multiple functions simultaneously.
4. One-shot learning
The one-shot learning approach is a variant of transfer learning in which we try to infer the required output from just one or a few training examples. It is useful in real-world scenarios where it is not feasible to label data for every possible category (if it is a classification task) or in situations where new categories can be added frequently.
5. Zero-shot learning
Zero-shot learning is another extreme variant of transfer learning, which relies on no labeled examples to learn a task.
What is the importance of Transfer Learning?
Transfer learning aims to save time and effort and provides the advantage of using tested models. This way, you cut costs by avoiding the need for a high-cost GPU for retraining the model. The goal is to make machine learning as human as possible.
What are the three theories of Transfer Learning?
The three theories of transfer learning are:
- Analogy
- Knowledge compilation
- Constraint violation
Each theory aims to predict human performance in distinct and identifiable ways on various transfer tasks.
What is Transfer Learning in Machine Learning?
Transfer learning is a machine learning method that uses a pre-trained model as the basis for a new model. Essentially, a model trained on one task is repurposed for a second related task to allow rapid progress when modeling the second task.
Examples of Transfer Learning
Transfer learning is used in various ways to strengthen machine learning models for natural language processing. Embedding pre-trained layers that understand specific dialects or vocabulary can be an example of simultaneously training a model to detect multiple elements of language.
Transfer Learning vs. Other Technologies & Methodologies
Transfer learning vs. fine-tuning
In transfer learning, a model developed for one task is used for another. In contrast, a fine-tuning approach to transfer learning involves changing the model output to fit a new task and training only the output model.
Transfer learning vs. domain adaptation
Domain adaptation is a subcategory of transfer learning. Domain adaptation involves the same feature space (but different distributions) for all domains; transfer learning includes cases where the target domain’s feature space differs from the source feature space.
Transfer learning vs. meta-learning
Meta-learning is more about speeding up and optimizing hyperparameters for not yet trained networks. In contrast, transfer learning uses a network that has already been trained and uses part of it to train on a new task that is relatively similar.
Transfer learning vs. reinforcement learning
Transfer learning involves fine-tuning a model trained on one set of data and then applying it to another collection of data and a different task. Reinforcement learning refers to how some agents should respond to environmental conditions to receive high rewards.
Transfer Learning in H2O AI
H2O Cascade (next-gen H2O Enterprise LLM Studio) fine-tunes pre-trained language models on your own data, a form of transfer learning.