Related papers: Comment on "Learning to Build the Bomb"
Would you like to have your own cryptography method? Experts say you should not do it. If you think you can develop a better cryptography method anyway. We present a brief discussion about some well known cryptography methods and how our…
Scientists who study how the brain solves problems have recently verified that, because of stringent limitations in working memory, where the brain solves problems, students must apply facts and algorithms that have previously been well…
In this article I describe a research agenda for securing machine learning models against adversarial inputs at test time. This article does not present results but instead shares some of my thoughts about where I think that the field needs…
A recent publication by the NSA assessing the usability of quantum cryptography has generated significant attention, concluding that this technology is not recommended for use. Here, we reply to this criticism and argue that some of the…
Distance bounding protocols are security countermeasures designed to thwart relay attacks. Such attacks consist in relaying messages exchanged between two parties, making them believe they communicate directly with each other. Although…
We study the outcomes of information aggregation in online social networks. Our main result is that networks with certain realistic structural properties avoid information cascades and enable a population to effectively aggregate…
AI that can accelerate research could drive a century of technological progress over just a few years. During such a period, new technological or political developments will raise consequential and hard-to-reverse decisions, in rapid…
The potential for advances in information-age technologies to undermine nuclear deterrence and influence the potential for nuclear escalation represents a critical question for international politics. One challenge is that uncertainty about…
"Fake news" is a recent phenomenon, but misinformation and propaganda are not. Our new communication technologies make it easy for us to be exposed to high volumes of true, false, irrelevant, and unprovable information. Future AI is…
There is growing concern over the potential misuse of artificial intelligence (AI) research. Publishing scientific research can facilitate misuse of the technology, but the research can also contribute to protections against misuse. This…
We consider global catastrophic risks due to cosmic explosions (supernovae, magnetars and gamma-ray bursts) and possible mitigation strategies by humans and other hypothetical intelligent beings. While by their very nature these events are…
The scientific method drives improvements in public health, but a strategy of obstructionism has impeded scientists from gathering even a minimal amount of information to address America's gun violence epidemic. We argue that in spite of a…
Potential advancements in artificial intelligence (AI) could have profound implications for how countries research and develop weapons systems, and how militaries deploy those systems on the battlefield. The idea of AI-enabled military…
"Machine unlearning" is a popular proposed solution for mitigating the existence of content in an AI model that is problematic for legal or moral reasons, including privacy, copyright, safety, and more. For example, unlearning is often…
The behavior of an atom in a molecule, liquid or solid is governed by the force it experiences. If the dependence of this vectorial force on the atomic chemical environment can be $learned$ efficiently with high-fidelity from benchmark…
Flaws and ambiguities are pointed out upon examining the comment attempting to solve a problem as raised recently --- the currently accepted formulation of electromagnetic radiation of an accelerated charge violates the principle of…
Short answer: Yes, Long answer: No! Indeed, research on adversarial robustness has led to invaluable insights helping us understand and explore different aspects of the problem. Many attacks and defenses have been proposed over the last…
Safe learning is central to AI-enabled robots where a single failure may lead to catastrophic results. Barrier-based method is one of the dominant approaches for safe robot learning. However, this method is not scalable, hard to train, and…
The last half-decade has seen a steep rise in the number of contributions on safe learning methods for real-world robotic deployments from both the control and reinforcement learning communities. This article provides a concise but holistic…
Although learning-based methods have great potential for robotics, one concern is that a robot that updates its parameters might cause large amounts of damage before it learns the optimal policy. We formalize the idea of safe learning in a…