We propose a prototype-based federated learning method designed for embedding networks in classification or verification tasks. Our focus is on scenarios where each client has data from a single class. The main challenge is to develop an embedding network that can distinguish between different classes while adhering to privacy constraints. Sharing true class prototypes with the server or other clients could potentially compromise sensitive information. To tackle this issue, we propose a proxy class prototype that will be shared among clients instead of the true class prototype. Our approach generates proxy class prototypes by linearly combining them with their nearest neighbors. This technique conceals the true class prototype while enabling clients to learn discriminative embedding networks. We compare our method to alternative techniques, such as adding random Gaussian noise and using random selection with cosine similarity constraints. Furthermore, we evaluate the robustness of our approach against gradient inversion attacks and introduce a measure for prototype leakage. This measure quantifies the extent of private information revealed when sharing the proposed proxy class prototype. Moreover, we provide a theoretical analysis of the convergence properties of our approach. Our proposed method for federated learning from scratch demonstrates its effectiveness through empirical results on three benchmark datasets: CIFAR-100, VoxCeleb1, and VGGFace2.

Large language models (LLMs) often benefit from intermediate steps of reasoning to generate answers to complex problems. When these intermediate steps of reasoning are used to monitor the activity of the model, it is essential that this explicit reasoning is faithful, i.e. that it reflects what the model is actually reasoning about. In this work, we focus on one potential way intermediate steps of reasoning could be unfaithful: encoded reasoning, where an LLM could encode intermediate steps of reasoning in the generated text in a way that is not understandable to human readers. We show that language models can be trained to make use of encoded reasoning to get higher performance without the user understanding the intermediate steps of reasoning. We argue that, as language models get stronger, this behavior becomes more likely to appear naturally. Finally, we describe a methodology that enables the evaluation of defenses against encoded reasoning, and show that, under the right conditions, paraphrasing successfully prevents even the best encoding schemes we built from encoding more than 3 bits of information per KB of text.

This is an edition of The Atlantic Daily, a newsletter that guides you through the biggest stories of the day, helps you discover new ideas, and recommends the best in culture. Yesterday, the OpenAI CEO Sam Altman testified before a Senate judiciary subcommittee about the "significant harm" that ChatGPT and similar generative-AI tools could pose to the world. When I asked Damon Beres, The Atlantic's technology editor, for his read on the hearing, he noted that Altman's emphasis on the broader existential risks of AI might conveniently elide some of the more quotidian problems of this new technology. I called Damon today to talk about that, and to see what else has been on his mind as he follows this story. Isabel Fattal: Can you talk a bit more about Altman's emphasis on the existential possibilities of AI, and what that focus might leave out?

Converging evidence has shown that human object recognition depends on familiarity with the images of an object. Further, the greater the similarity between objects, the stronger is the dependence on object appearance, and the more important two(cid:173) dimensional (2D) image information becomes. These findings, how(cid:173) ever, do not rule out the use of 3D structural information in recog(cid:173) nition, and the degree to which 3D information is used in visual memory is an important issue. Liu, Knill, & Kersten (1995) showed that any model that is restricted to rotations in the image plane of independent 2D templates could not account for human perfor(cid:173) mance in discriminating novel object views. We now present results from models of generalized radial basis functions (GRBF), 2D near(cid:173) est neighbor matching that allows 2D affine transformations, and a Bayesian statistical estimator that integrates over all possible 2D affine transformations.

Machine Learning, a sub-branch of Artificial Intelligence, has established itself as the new go-to technology for businesses worldwide. Whether it is e-commerce or healthcare, almost all the industries are using Machine Learning extensively to make futuristic solutions and products. Machine Learning depends heavily on programs and algorithms that help machines self-learn without having to be instructed explicitly. Machine Learning is pretty much dictating our daily lives- how, you wonder? Let's look at the top applications of Machine Learning to understand how it is shaping the digital economy.

So far at #CES2020, it's very obvious that companies will almost always focus on how AI can help people do and consume more. It's never like, "Oh, AI figured out how to handle a lot of stuff. Let's work less and consume less and chill out."

The explosion of digital has caused a shift in customer buying expectations. Customers can easily educate themselves about products, suppliers, and competitive pricing with access to more information than ever before. Many businesses struggle to grow revenue and increase wallet share because of ineffective strategies to penetrate their existing accounts. With the acquisition cost for a new customer at 5x more than an existing customer, and existing customers spending 31% more on average, it's essential that businesses empower sales with the ability to make personalized, relevant product recommendations to their current customers. If you've invested in sales tools (like PROS Smart CPQ) – then you've already taken proactive steps in the right direction.

Ironically, people are more likely to trust AI when they don't know it's being used. AI is proving itself superior to human intelligence in an expanding number of fields. That is, except when people know AI is being used. Yes, because in certain human-centric sectors, the performance of artificial intelligence starts to drop off if people are apprised of the involvement of an intelligent machine. In fact, human resistance would seem to be the achilles heel of artificial intelligence, since for all the recent advances of AI technology this resistance is preventing AI from doing its job in areas where human contact and interaction would normally play a central role. This message was brought home most recently by a study published in Marketing Science on September 20, titled "The Impact of Artificial Intelligence Chatbot Disclosure on Customer Purchases."

Ironically, people are more likely to trust AI when they don't know it's being used. AI is proving itself superior to human intelligence in an expanding number of fields. That is, except when people know AI is being used. Yes, because in certain human-centric sectors, the performance of artificial intelligence starts to drop off if people are apprised of the involvement of an intelligent machine. In fact, human resistance would seem to be the achilles heel of artificial intelligence, since for all the recent advances of AI technology this resistance is preventing AI from doing its job in areas where human contact and interaction would normally play a central role. This message was brought home most recently by a study published in Marketing Science on September 20, titled "The Impact of Artificial Intelligence Chatbot Disclosure on Customer Purchases."

As hot as stories about artificial intelligence (AI), augmented/virtual reality (AR/VR), blockchain, and the Internet of Things (IoT) have been in recent months, we often can't help but think of these technologies as a long way off from mainstream adoption. Movies like Ready Player One and Avengers: Infinity Wars only perpetuate this perception by mixing real technologies with fantasy, making the tech we wield in the real world seem primitive in the process. AI, AR/VR, blockchain, and IoT are already playing an important role in our everyday lives, with countless examples hiding in plain sight. In this first article, we're going to look at Artificial Intelligence (AI). In pop culture, AI is often represented as a highly-evolved and self-aware artificial "brain" in a robot capable of overthrowing the human race, as in Ex Machina or The Terminator movies.