News for hackers. Stuff that matters.
Re:The Beginner's Textbook for Homomorphic Encryption by EGreg on Sunday September 21, 2025 @06:07PM (#45327117)
Funny thing is

Since neural networks are differentiable, they can be homomorphically encrypted!

That’s right, your LLM can be made to secretly produce stuff hehe

Re:The Beginner's Textbook for Homomorphic Encryption by LeGrosDadai on Sunday September 21, 2025 @10:52PM (#45328669)
That's pretty cool, but isn't any computable function can be computed via FHE, so I'm not sure the differentiable part is necessary.
Re:The Beginner's Textbook for Homomorphic Encryption by ogogmad on Tuesday September 23, 2025 @08:56AM (#45346399)
Any program which you apply FHE to needs to be expressed as a circuit, which implies that the time taken to run a computation needs to be fixed in advance. It's therefore impossible to express a branch instruction (or "if" statement, if you prefer).

The circuits are built out of "+" and "×" gates, which are enough to express any polynomial. In turn, these are enough to approximate any continuous function (Weierstrass's approximation theorem). In turn, every computable function on the real numbers is a continuous function - so FHE is very powerful.

Re:The Beginner's Textbook for Homomorphic Encryption by jlokier on Tuesday September 23, 2025 @11:51AM (#45348750)
> In turn, every computable function on the real numbers is a continuous function

That doesn't seem right. Consider the function f(x: ℝ) = 1 if x >=:0, 0 otherwise. That's computable but not continuous.

Re:The Beginner's Textbook for Homomorphic Encryption by ogogmad on Tuesday September 23, 2025 @03:54PM (#45351941)
That's uncomputable because equality of real numbers is undecidable. Think infinite strings of digits.
Re:The Beginner's Textbook for Homomorphic Encryption by gametorch on Monday September 22, 2025 @02:32PM (#45337583)
ReLU, commonly used in neural networks, is not differentiable at zero but it's still able to be approximated by expressions that are efficiently FHE-evaluable. You don't truly care about differentiability here, if you're being pedantic.

Very insightful comment, though. LLMs run under FHE (or just fully local LLMs) are a great step forwards for mankind. Everyone should have the right to interact with LLMs privately. That is an ideal to strive for.

Re:The Beginner's Textbook for Homomorphic Encryption by seanhunter on Monday September 22, 2025 @03:49PM (#45338564)
Differentiability isn’t a requirement for homomorphism I don’t think.

Homomorphism just means say I have a bijective function [1] f: A -> B and a binary operator * in A and *’ in B, f is homomorphic if f(a1*a2) = f(a1)*’f(a2). Loosely speaking it “preserves structure”.

So if f is my encryption then I can do *’ outside the encryption and I know because f is homomorphic that the result is identical to doing * inside the encryption. So you need your encryption to be an isomorphism [2]and you need to have ”outside the encryption “ variants of any operation you want to do inside the encryption. That is a different requirement to differentiability.

1: bijective means it’s a one to one correspondence

2: a bijection that has the homomorphism property is called an isomorphism because it makes set A equivalent to set B in our example.