Professional Documents
Culture Documents
Lesson 1
With your groupmates, choose at least one among the following articles or cases:
Author/s:
Summary Narrative:
Internet of Things (IoT) is a network of physical devices embedded with sensors, software, and
connectivity that can communicate over the network with other interconnected devices. DL has been
extensively applied in a variety of application domains such as speech recognition, medical diagnosis,
but the recent security and privacy issues of DL have raised concerns of the researcher. One of the keys
to the rise of DL is to rely on the vast quantities of data, which is also accompanied by the risk of privacy
leakage. In this paper, we first describe the potential risks of DL and then reviewed the two types of
attack: model extraction attack and model inversion attack in DL and four typical defense technologies
for protecting the data privacy of user: DP, HE, SMC, and TEE. We then investigated two types of attacks:
adversarial attacks and poisoning attacks. In adversarial attacks, we reviewed the representative black
box and white box attack in recent studies, and reviewed the adversarial attacks under the physical
condition. Regarding the defense methods of security, we describe the defense approach from three
aspects: pre-processing, improving model robustness, and malware detection. Finally, the unresolved
problems and the direction of future work are discussed.
Deep learning has now become part of our daily lives, and when new technology invested, definitely
security and pri-vacy issues arise. In recent years, extensive research was carried out on the security and
privacy preserving issues and it's counter frameworks for Deep Learning and Deep NeuralNetwork’s
training and interface modules. Therefore, security and privacy become very critical and important
issues as in the other technologies that cannot be overlooked.
•Deep learning is a method in artificial intelligence (AI) that teaches computers to process data in a way
that is inspired by the human brain. Deep learning models can recognize complex patterns in pictures,
text, sounds, and other data to produce accurate insights and predictions. You can use deep learning
methods to automate tasks that typically require human intelligence, such as describing images or
transcribing a sound file into text.
In using deep learning method our suggestion is to share less information with apps and services so that
we can avoid any problems. Also, we must use strong and uniques password.this make a sense when it
comes to identification of individuals, unauthorized commercial sharing of confidential information,
illegitimate use of private data, and the disclosure of sensitive data or inferred private information.
Group Members:
AALA, ROSARIO F.
ALMOITE, ROSEMEN
BABOR, EDELYN
BABOR, MAY
BARAERO, MICAELLA
BENDECIO, NOELA
BERMAS, CHARLENE
BERMAS, JANICE