跳到主要内容
Learn Motion
探索课程产品能力价格

下载 Learn Motion

用手机扫描二维码,下载 App。

App Store

扫码从 App Store 下载 Learn Motion打开商店页面

Google Play

扫码从 Google Play 下载 Learn Motion打开商店页面
登录
Learn Motion

How AI Is Changing Cancer Drug Discovery

1Why Cancer Drug Discovery Is So Hard2What AI Can and Cannot Do Here3Finding the Right Target4Designing Molecules with AI5Testing, Trials, and Real-World Impact6What's Next and What to Watch
What AI Can and Cannot Do Here

Why the Data Sets the Ceiling

2 / 3
Think of the scale as a ceiling, not a balance you can tip by trying harder. Three separate things about the data push on it. How much data there is: with only a few examples, a pattern may just be coincidence. How good the labels are: if the same experiment was run under different conditions, the model learns the inconsistency as if it were biology. And who is in the data: a model trained on one cancer subtype will quietly lose accuracy on another. The last one is what bias looks like in practice — not an abstraction, just a measurable drop when the model meets patients it never saw.
0:00 / 0:00

A model's accuracy is bounded by its training data in three separate ways, and it is worth keeping them apart.

Quantity: patterns that appear in a handful of examples may be coincidence. With enough examples, the model can distinguish a real regularity from noise. Cancer datasets are often small in exactly the categories that matter most — rare tumor types, unusual patient populations — so the model is weakest where the need is greatest.

Quality: labels are only as trustworthy as the measurement behind them. If activity was assayed under inconsistent lab conditions, or if patient records code the same outcome in different ways, the model learns those inconsistencies as if they were biology.

Representativeness: a model trained mostly on one population or one cancer subtype will perform poorly outside it. This is the bias problem in concrete form — not a moral abstraction, but a measurable drop in accuracy when the model meets patients who were never in its training set.

A useful way to hold this together: the data defines the ceiling, and no amount of clever modeling raises it.

上一页2 / 3下一页

Learn Motion

生成课程,认真学会。

运营方:Wuhan Daoyin Technology Co., Ltd.

联系我们: [email protected]
隐私政策用户服务协议

© 2026 Learn Motion