Questions tagged [machine-learning]
Machine learning is an application of artificial intelligence (AI) that provides systems the ability to automatically learn and improve from experience without being explicitly programmed. Machine learning focuses on the development of computer programs that can access data and use it learn for themselves.
31
questions
1
vote
0
answers
21
views
Non-Gaussian distribution in continuous learning with error
The CLWE problem (and related) talks about the hardness of finding the secret key $\vec{s}$, given polynomially many samples $(\vec{a},t)$, where $\vec{a}$ is sampled from the normal distribution, and ...
- 111
1
vote
0
answers
14
views
Gradient Leakage Attacks in Federated Learning
May I ask if I use top-K to compress the gradient, can the attacker recover the original information of the data from the compressed gradient?
- 41
2
votes
1
answer
85
views
Key management problem in federated learning based on homomorphic encryption
In federated learning using homomorphic encryption, all participants in most schemes share the same pair of keys, which can easily cause key leaks and lead to data privacy leaks.
After research, I ...
- 41
0
votes
0
answers
36
views
A Method to Preserve Gradient Privacy in Federated Learning
In the federated learning architecture, there are two methods of gradient privacy protection: differential privacy and homomorphic encryption. What are the advantages and disadvantages of these two ...
- 41
2
votes
2
answers
377
views
Can ML be used to overcome cryptography
I saw some recent papers(e.g Encrypted DNS --> Privacy? A Traffic Analysis Perspective) about adopting ML technology to overcome cryptography implemented to ensure network security. Network packets ...
- 133
1
vote
1
answer
194
views
Binary Secret Sharing vs Garbled Circuits
In Privacy-preserving machine learning, GC is usually used for privacy operation such as ReLU(x) where sign(x) needs to know. However, binary secret sharing also supports such computation via ...
- 11
0
votes
0
answers
87
views
Cryptographic functions as feature map/kernel function?
Has there been any use of cryptographic function as a kernel function with support vector machine? There are several standard kernels to be used with SVMs each with its own scenario.
I was not able to ...
2
votes
0
answers
59
views
Using ML to detect what classical cipher the ciphertext is encrypted with
I was considering creating an ML project where it is fed some ciphertext by any classical cipher and would return possible ciphers that encrypted the text. I would have to create a sizeable dataset ...
- 21
-1
votes
1
answer
99
views
Neural Network based on pseudorandom number
Recently, I read this paper NEURAL NETWORK BASED
CRYPTOGRAPHY. Under the section 3.1 it said:
The aim is to improve the randomness of the random numbers generated by any
algorithm using an NN. In ...
- 99
0
votes
1
answer
73
views
Adaptive security
I would like to know if there are any effective methods to choose a cryptographic algorithm from a pool of algorithms depending on a certain situation.
Let's say we know the performances of these ...
- 1
0
votes
1
answer
109
views
Extracting genome from a Ciphertext
Cryptography is the practice of securing communication from unauthorized access. One way of securing information is through encryption, where plaintext (clear text) is transformed into ciphertext ...
- 1,930
0
votes
1
answer
52
views
security of different federated learning schemes
all, I am working on federated learning and here goes my question:
Suppose there are two participants to do the federated learning. For some of the models (e.g. Logistic Regression models, assume one ...
- 149
-3
votes
1
answer
319
views
Can Big Data together with deep neural networks attack RSA by affording the vast calculation of prime multiplications in advance?
This is a spin-off from Can Big Data attack RSA by just calculating many prime multiplications in advance? [duplicate].
Intro
I am somewhat new to cryptography.
Repeating the basics of RSA from How ...
- 99
0
votes
0
answers
144
views
Paillier scheme : Encoding floats into integers impact on computations
In Privacy Preserving Processing Over Encrypted Images, I could understand that appropriate encoding of floats into integers (required in Paillier) only incur negligible error in computations.
Any ...
- 75
2
votes
1
answer
102
views
How to adapt the equation of Gaussian mechanism noise based on number of executions
I'm trying to build a differentially private machine learning model. I'm using the Gaussian mechanism to calculate the required noise amount based on pre-defined privacy budget value 𝜖
The equation ...
- 23
2
votes
1
answer
57
views
Selection of the noise application position in differential privacy
In DP-SGD proposed by M Abadi in 2016, noise is applied to the gradient, so every round of training needs to be applied. My questions are:
Can I choose to apply noise that meets the DP requirements to ...
- 31
2
votes
0
answers
52
views
Predicting with a machine learning model while preserving the privacy
Imagine Alice has trained a machine learning model. She wants to store her model in a blockchain so that everyone can use it; however, she wants her model to be private so that no one can steal her ...
- 21
4
votes
1
answer
216
views
Zero knowledge proof for verifying a machine learning model
Imagine Alice has trained a machine learning model. Bob wants to verify that whether Alice actually trained the model or not, but Alice does not want to reveal her model (because the model is personal ...
- 65
0
votes
1
answer
202
views
Machine Learning to improve Encryption algorithms [duplicate]
Can machine learning be used to improve encryption algorithms e.g.image encryption algorithms that use ECC ?
0
votes
0
answers
31
views
Predict with a machine learning model without knowing the actual model values
Imagine Alice has trained a machine learning model. She wants other people like Bob to be able to use her model for prediction, but she doesn't want them to know the actual model. She wants the model ...
- 65
1
vote
0
answers
70
views
The existing approaches based on machine learning for cryptography [closed]
i'm working on a paper about Machine learning and Deep Learning and i'm wondring about the uses of this domain in cryptography!!
so what are these application of Ml that we use in cryptography and the ...
0
votes
1
answer
52
views
Decrypting data on client side for use inside program
So I am developing a python program and encrypting the trained machine learning models which need to be decrypted when loading them on client side. My program will deployed on the client's machine ...
0
votes
0
answers
367
views
Computing guessing entropy (average rank of the correct key) in attack using deep learning
I am looking for help with some issues that I have about the guessing entropy (GE) in side-channel attacks context. I have read some papers where GE is used as a metric to measure the attack's ...
- 11
1
vote
2
answers
130
views
What machine learning accuracy is assumed to be predictable for TRNG/PUF application?
In regular machine learning (ML) applications, usually an accuracy of greater than 95% is desired.
In an ideal TRNG/PUF applications, unpredictable behavior (50% accuracy with ML models) is desired. ...
- 263
2
votes
1
answer
322
views
Applications of machine learning in classical ciphers?
Machine learning is definitely applicable in analyzing simple shift ciphers like Caesar and affine ciphers, as well as substitution ciphers like Vigenère, but is it possible for machine learning to ...
1
vote
0
answers
703
views
how to measure/calculate guessing entropy for side channel attacks
I was going through the paper (https://link.springer.com/chapter/10.1007/978-3-319-66787-4_3), and I came across where authors says that accuracy alone is not enough as an evaluation metric and ...
- 11
6
votes
5
answers
3k
views
Using AI to perform Cryptanalysis
If given a large set of examples of cyphertext and corresponding plaintext, could AI be trained to decrypt a cyphertext as the examples provided?, and if so, are there any examples online ...
- 183
0
votes
1
answer
116
views
Can machine learning/AI lower the security existing cryptographic protocols similar to quantum computers? [duplicate]
Initial thinking would be not unless the protocol revealed sensitive local information that could then be analysed by an ML algorithm and applied globally.
local meaning I see messages from Alice and ...
- 1,303
2
votes
1
answer
136
views
How can we compare an encrypted number with a normal number?
I am currently doing a project on privacy preserving encryption using a k-means system and the Paillier encryption algorithm (homomorphic algorithm). I have to send an image of a skin disease to the ...
8
votes
1
answer
4k
views
Deep Learning application in decryption?
If the output of an algorithm when interacting with the [encryption] protocol matches that of a simulator given some inputs, it ‘need not know’ anything more than those inputs.
Can a machine learn to ...
- 1,930
6
votes
1
answer
3k
views
Homomorphic Encryption for Deep Learning
I'm interested in the two following processes:
Perform deep learning on homomorphic encrypted data
Perform deep learning predictions with a homomorphic encrypted model on unencrypted data. By this, ...
- 313