Naru News日本語

Home › Glossary › AI

What is transfer learning in AI?

Term explainer · Checked November 18, 2026 · 11 min read

In one minute

Transfer learning is a way for an AI to put what it learned on one task to use when it learns a different task.

For example, an AI that practiced naming objects in more than a million photos can become the starting point for an AI that spots diseases in a small set of medical images. The early layers (stages of calculation) of an image AI have learned clues that help with almost any photo, such as lines and patterns. Borrowing them means far less data is needed than learning from scratch.

There are two common ways to reuse them: freeze the borrowed layers and train only a new last layer, or also retrain the borrowed layers a little (fine-tuning). The stage that builds the starting point from huge amounts of data is pre-training. But if the old and new tasks are too different, results can get worse instead (negative transfer).

A common mix-up

“Transfer learning and fine-tuning are two separate methods.”

Only partly right. Transfer learning is the name of an idea: putting what was learned on one task to use on another. There are two main ways to do it. One is to freeze the learned layers so they don’t change and train only a newly added last layer (it borrows just the clues, or “features,” the learned layers pull out, so this is also called feature extraction). The other is to start from the learned numbers (parameters) and also retrain the borrowed layers little by little on the new data; that is fine-tuning. It is true that these two ways differ. But fine-tuning is not a separate technique outside transfer learning. Machine learning textbooks and the guides for AI-building tools also describe it as one of the common ways to do transfer learning. Which way to choose depends on how much data you have and how close the new task is to the old one. In fine-tuning, how much the numbers inside are changed at each learning step (the learning rate) is kept very small so the borrowed numbers aren’t ruined.

Explained in pictures

How an image AI’s learned layers are borrowed to learn a new task (an example made for this page). Title: “Borrow learned layers to learn a new task.” Top row, “① Build the starting point (pre-training)”: a stack of photos, “Over a million photos” → three boxes, “Lines, colors,” “Patterns,” “Parts” → a navy box, “Output layer: 1,000 names,” with “Long training on lots of data” to the right. Under the three boxes: “Early layers: clues useful for any photo.” A dotted arrow down: “Copy the learned numbers as they are.” Bottom part, “② Reuse for a new task” (e.g., spotting disease in medical images), in two rows. “Way 1: Borrow features only”: the three copied boxes with a padlock mark on top, then an orange box, “New output: disease or not,” and “Only the output learns.” “Way 2: Fine-tuning”: the three copied boxes with a turning-arrow mark on top, the same “New output,” and “Borrowed layers adjust a bit too.” Band at the bottom: Start from learned numbers, not scratch, so less data is needed.
The top row is pre-training, which builds the starting point: the AI practices naming 1,000 kinds of objects in more than a million photos. The early layers (boxes) learn clues useful for any photo, such as lines, colors, patterns and parts. The output layer on the far right is tied to the old task’s 1,000 names, so it isn’t used for the new task. In the bottom part, the numbers of the early layers are copied as they are, a new output layer is added, and the AI learns from a small set of medical images. Way 1 freezes the borrowed layers (the padlock mark) and trains only the new output. Way 2, fine-tuning, also adjusts the borrowed layers little by little (the turning-arrow mark).
What each layer of an image AI responds to (from papers in 2013 and 2014). Title: “Early layers help other tasks most.” Six boxes joined by arrows, left to right. “Layer 1”: stripes at different angles and an orange dot, “Edge angles, color blobs.” “Layer 2”: corner shapes, “Corners, edge + color.” “Layer 3”: a mesh, “Mesh-like patterns.” “Layer 4”: a dog’s face, “Parts like dog faces.” “Layer 5”: a keyboard, “Whole objects like keyboards.” A navy “Output” box: “Gives 1,000 names.” Below, a double-headed arrow labeled “Shared by many tasks (easy to borrow)” at the left end and “Specific to the old task (hard to borrow)” at the right end. A box below: “Similar tasks: Borrowing even late layers costs little.” “Different tasks: The more late layers borrowed, the bigger the drop.” Band at the bottom: Early layers end up alike, whatever photos they learn from.
This figure lines up, from left to right, what each layer of an image AI responds to. Layer 1 responds to simple things such as the angle of lines and blobs of color. These clues come out much the same whatever photos the AI learns from, so they are easy to borrow for other tasks. Later layers respond to more complex things tailored to the old task: corners, patterns, parts and whole objects. The output layer is the old task’s answer itself. In a 2014 study, borrowing even late layers cost little between similar tasks, but between different tasks, such as man-made objects and natural things, the more late layers were borrowed, the more the results dropped (the box at the bottom).
Horizontal bars comparing accuracy on a task of telling apart photos of 101 kinds of objects (figures from a 2013 paper; Caltech-101). Title: “The fewer the photos, the more borrowing helps.” A scale at the top: 0%, 50% and 100%. “Learning from 15 photos per kind”: “From scratch,” a short gray bar, 22.8%; “Borrowed layers,” a long orange bar, 83.8%. “Learning from 30 photos per kind”: “From scratch,” a gray bar, 46.5%; “Borrowed layers,” a long orange bar, 86.5%. A note below: Both AIs have the same design. The “Borrowed layers” AI kept layers learned from over a million photos frozen and trained only a new output layer. Band at the bottom: The fewer the photos, the bigger the gap with learning from scratch.
This figure compares how often AIs named the right object on a task with 101 kinds of objects (figures from a 2013 paper). The gray bars are an AI that learned from scratch on that task’s photos alone; the orange bars are an AI that borrowed layers learned from more than a million photos, kept them frozen, and trained only a new output layer. The two have the same design. Learning from 15 photos per kind, the AI that started from scratch got only 22.8% right, while the one with borrowed layers got 83.8%. With 30 photos per kind, the AI from scratch rose to 46.5% and the gap narrowed. Borrowed clues help most when photos are few.

You’re free to use these figures in lessons, slides and posts if you credit Naru News (naru.news). CC BY 4.0.

Think of it like this

Someone who played the violin learns the piano faster

Someone who has learned the violin doesn’t have to start over on things like reading music when they begin the piano. Both are musical instruments, so they share some knowledge. What you learned before helps you learn the next thing: that is the idea behind transfer learning. In an AI, the counterpart of reading music is the clues, such as lines and patterns, that its early layers have learned.

But not everything helps. Learning to ride a bicycle won’t make you better at the piano any faster. In AI, if the old and new tasks are too different, the borrowed clues may not just fail to help but make results worse than not borrowing at all (negative transfer).

The comparison has limits, too. A person works out for themselves which experience might help, but with an AI, the people building it choose which AI and which layers to borrow. And for an AI, “borrowing” means copying the learned numbers as they are to use as the new AI’s starting point. Because they are copied, the original AI stays as it was.

Key points

Why is it called “transfer”?

The word comes from psychology, the study of how people learn. When something learned earlier affects later, different learning, it is called “transfer of learning.” When it helps the later learning, it is “positive transfer”; when it gets in the way, it is “negative transfer.”

Transfer learning in machine learning is thought to come from the same idea. Just as someone who has learned the violin can learn the piano faster, an AI that puts what it learned before to use should be able to learn a new task well with less data.

One machine learning glossary describes it as transferring information from one task to another, for example, from a task with plenty of data to one with little. Collecting data with answers attached takes time and money, so needing less of it is a big advantage.

Why does what was learned help with another task?

In AIs that recognize images, the early layers learn simple clues that show up in any photo. It is known that in deep networks (calculating systems with many stacked layers) trained on photos, the first layer comes to respond to stripes at various angles and to blobs of color, whatever photos it learned from. Because much the same thing forms whatever the task, these clues are thought to be shared ones that other tasks can use too.

The later the layer, the more its clues are tailored to the original task. In a 2013 study, the second layer responded to corners, the third to mesh-like patterns, the fourth to parts such as dogs’ faces, and the fifth to whole objects such as dogs and keyboards (more on the “Deep learning” page). The last, output layer gives the original task’s answers (such as 1,000 names), so it is replaced for a new task.

A 2014 study checked this layer by layer. It split the 1,000 kinds in more than a million photos into two tasks of 500 kinds each, copied the first few layers of a network trained on one task, and then trained on the other. The first two layers could be borrowed with almost no loss. When the split made the tasks different, such as man-made objects versus natural things, the more later layers were borrowed, the bigger the drop.

How is it done?

There are two common ways. Both begin by copying the layers and numbers of a trained AI and swapping only the last, output layer for one made for the new task. The old output layer is tied to the old task’s answers (such as 1,000 names).

The first way freezes the borrowed layers so they don’t change and trains only the new output layer (borrowing just the features, or feature extraction). The AI learns only how to connect the clues the borrowed layers pull out of a photo to the new answers. The layers are frozen so that retraining won’t ruin the information they hold.

The second way, done afterward, unfreezes all or some of the borrowed layers and retrains them on the new data too; that is fine-tuning. The clues are adjusted little by little to fit the new task. But the new data is often small, so changing the numbers inside by a lot at each learning step (a high learning rate) can quickly lead to overfitting (fitting the training data so closely that the AI fails on real cases).

The guide for an AI-building tool has an example. An AI learns to tell cats from dogs using 2,000 photos. It borrows a network trained on more than a million photos, keeps it frozen, and trains only a newly added output layer; its accuracy reached about 96%. Then the later layers were unfrozen and fine-tuned, and accuracy rose to nearly 98%.

How big a difference does it make when data is scarce?

The gap can be very large. The 2013 study above, which looked at what each layer responds to, also tested how much borrowing helps, in the same paper. It borrowed the layers of a network trained on more than a million photos, kept them frozen, trained only a new output layer, and had it tell apart photos of 101 kinds of objects. Learning from 15 photos per kind, it got 83.8% right. The same network design, trained from scratch on just those photos, got 22.8%. With 30 photos per kind, the figures were 86.5% and 46.5%.

It is used for medical images too. A 2017 study started from an image AI trained on huge numbers of photos and fine-tuned all its layers on about 130,000 photos of skin conditions. On two tasks of telling skin cancers from similar non-cancerous spots (such as moles), it did as well as 21 board-certified dermatologists.

How do language AIs use it?

Transfer learning took off in language AI around 2018. A paper that year pointed out that transfer learning had transformed the image field, but in language, models were still often trained from scratch for each task. It then showed a method that fine-tunes a language model trained on huge amounts of text. On tasks of sorting texts into types, with only 100 labeled examples, it did as well as a model trained from scratch on 100 times as many.

In the image field, pre-training usually meant learning from huge numbers of labeled photos. In language, pre-training on text without answers attached became common. If the AI practices guessing the next word or a hidden word, the answer is already written in the text itself (more on the “Pre-training” page).

How is it different from pre-training and fine-tuning?

Transfer learning is the name of an idea: “put what was learned to use on another task.” Pre-training and fine-tuning are names for a stage and a method within it.

Pre-training is the stage that builds the starting point. The AI learns in advance, from huge amounts of data, a foundation that helps with many tasks. For images, this often means huge numbers of labeled photos; for language, huge amounts of text without answers attached.

Fine-tuning is one way of reusing it. Starting from the borrowed numbers, the AI is retrained little by little on data for the target task. The other way is to freeze the borrowed layers and train only the output layer (more on the “Fine-tuning” page).

Transfer learning also takes forms other than borrowing layers. Having one model handle several tasks (multi-task learning) is given as an example of transfer learning.

Can it go wrong?

Yes. If the old and new tasks are too different, borrowing can make the results worse than not borrowing at all. This is called “negative transfer.” In machine learning, a 2005 study found it by experiment and concluded that transfer learning can actually hurt performance when tasks are too dissimilar. The main cause is thought to be the difference between the old data and the new data. It can also happen when the old data has many errors or much noise.

The bigger the difference, the less there is to gain from borrowing. In the 2014 study above, which checked layer by layer, between different tasks (man-made objects versus natural things), borrowing later layers caused a much bigger drop than between similar tasks.

With plenty of data, borrowing may make no difference. A 2019 study reported that on medical image tasks with plenty of data, such as about 250,000 photos of the back of the eye, borrowing made almost no difference to results. But with the photos cut to 5,000, borrowing gave the large models better results.

Check yourself

  1. Q1The many stacked layers of an AI trained to recognize photos are borrowed for a different photo task. Which layers tend to help other tasks as they are?

    Answer and explanation

    Answer: A The early layers, near the input

    The early layers near the input come to respond to clues found in any photo, such as stripes at various angles and blobs of color, whatever photos they learned from. Moving back from there, the clues become more tailored to the old task, and the last, output layer is tied to the old task’s answers, so it is replaced for a new task. In a 2014 study, too, the first two layers could be borrowed for another task with almost no loss.

    More on this → Key points: Why does what was learned help with another task?

  2. Q2Which describes how “transfer learning” and “fine-tuning” in AI are related?

    Answer and explanation

    Answer: C Fine-tuning is a kind of transfer learning

    Transfer learning is the name of the idea of putting what was learned to use on another task, and fine-tuning is one way of doing it: starting from the borrowed numbers, the AI is retrained little by little on the new task’s data. Freezing the borrowed layers and training only a new output layer is another way of doing transfer learning, so the two are not the same thing.

    More on this → Key points: How is it different from pre-training and fine-tuning?

  3. Q3A 2014 study borrowed the learned layers of a photo-recognizing AI between two tasks: photos of man-made objects and photos of natural things. When more layers, further toward the back, were borrowed, what happened to the new task?

    Answer and explanation

    Answer: B The more layers borrowed, the worse it did

    Between different tasks such as man-made objects and natural things, the more of the later layers were borrowed, the bigger the drop in results. This is because the later the layer, the more its clues are fitted to the original task. If the old and new tasks are too different, the new task can even do worse than without borrowing. This is called negative transfer.

    More on this → Key points: Can it go wrong?

Sources and how we checked

AI wrote this page from the sources above, and a different AI checked it against the text of those sources (November 18, 2026). A person with a master’s degree in science and a Japanese high-school science teaching license also read and checked it before publishing. Numbers are as of that date. If anything was hard to follow, you spot a mistake, or there’s a term you’d like explained, please write to [email protected].

AI terms

All 58 AI terms →