基于正规化视角的理论分析,对深度尖端神经网络的对抗性强度进行了理论分析
Hui Zhang1, Jian Cheng2, Jun Zhang1
1Nanjing University of Science and Technology, Nanjing, 210094, China.
概括
训练尖端神经网络 (SNN) 通过引入一个调节器来最大限度地降低输出梯度,直接提高对手的稳定性. 这使得SNN与转换的ANN相比,更适应攻击.
科学领域:
- 人工智能的人工智能
- 计算神经科学是一种神经科学.
- 机器学习 机器学习
背景情况:
- 尖端神经网络 (SNN) 是第三代神经网络.
- 将预先训练的人工神经网络 (ANN) 转换为SNN是计算效率高的,但导致对敌对攻击的脆弱性.
- 虽然直接培训提高了SNN的稳定性,但基础的理论机制仍然不清楚.
研究的目的:
- 为直接受过训练的SNNs的对抗性稳定性提供理论解释.
- 分析预期的风险功能,并确定有助于稳健性的机制.
- 通过经验验证来证明这种方法的有效性.
主要方法:
- 在SNN中建模Poisson编码器的随机过程.
- 对预期风险函数的理论分析,以确定一个积极的半确定的调节器.
- 在CIFAR10和CIFAR100数据集上进行广泛的实验,以验证理论发现.
主要成果:
- 一个理论证明揭示了训练有素的SNN中一个正的半确定的调节器.
- 这种调节器有效地减少了输出与输入的梯度,提高了强度.
- 实验结果显示,与转换的SNN相比,受过训练的SNN的梯度大小明显较小 (13-160倍小).
结论:
- 通过直接训练SNN,通过特定的调节器引入了固有的强度机制.
- 将梯度的平方和最小化是提高SNN对抗性弹性的关键.
- 这些发现为开发更强大的SNNs来应对对手攻击提供了理论基础.
相关概念视频
Neural Regulation
39.6K
Digestion begins with a cephalic phase that prepares the digestive system to receive food. When our brain processes visual or olfactory information about food, it triggers impulses in the cranial nerves innervating the salivary glands and stomach to prepare for food.
39.6K
Neural Circuits
1.3K
Neural circuits and neuronal pools are two of the main structures found in the nervous system. Neural circuits are networks of neurons that work together to carry out a specific task or process. They consist of interconnected neurons and glial cells, which provide structural and metabolic support.
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
1.3K
Generalization, Discrimination, and Extinction
658
Generalization, discrimination, and extinction are key concepts in operant conditioning that influence how behaviors are learned and maintained.
Generalization occurs when a behavior reinforced in one context is performed in similar situations. For instance, a student who studies diligently for calculus and receives excellent grades might apply the same study habits to psychology and history, expecting similar results. Generalization shows how learning in one setting can influence behavior in...
Generalization occurs when a behavior reinforced in one context is performed in similar situations. For instance, a student who studies diligently for calculus and receives excellent grades might apply the same study habits to psychology and history, expecting similar results. Generalization shows how learning in one setting can influence behavior in...
658
Survival Tree
119
Survival trees are a non-parametric method used in survival analysis to model the relationship between a set of covariates and the time until an event of interest occurs, often referred to as the "time-to-event" or "survival time." This method is particularly useful when dealing with censored data, where the event has not occurred for some individuals by the end of the study period, or when the exact time of the event is unknown.
Building a Survival Tree
Constructing a...
Building a Survival Tree
Constructing a...
119
Stability of structures
200
In mechanical engineering, the stability of systems under various forces is critical for designing durable and efficient structures. One fundamental way to explore these concepts is by analyzing systems like two rods connected at a pivot point, O, with a torsional spring of spring constant k at the pivot point. This system is similar in appearance to a scissor jack used to change tires on a car. In this case, the arms of the linkage (equivalent to the rods in this system) are entirely vertical,...
200
Propagation of Action Potentials
6.0K
The propagation of an action potential refers to the process by which a nerve impulse, or "action potential," travels along a neuron.
Neurons (nerve cells) have a resting membrane potential, with a slightly negative charge inside compared to outside. This is maintained by ion channels, such as sodium (Na+) and potassium (K+) channels, which control the flow of ions. When a stimulus, like a touch or a signal from another neuron, triggers the neuron, sodium channels open, allowing sodium ions to...
Neurons (nerve cells) have a resting membrane potential, with a slightly negative charge inside compared to outside. This is maintained by ion channels, such as sodium (Na+) and potassium (K+) channels, which control the flow of ions. When a stimulus, like a touch or a signal from another neuron, triggers the neuron, sodium channels open, allowing sodium ions to...
6.0K


