These plant-based meal kits and delivery services bring healthy preprepared meals and meal kits to your door. Plant-Based meal kit services are a modern miracle for vegetarians and vegans, who usually aren't afforded the same conveniences as meat eaters or those without dietary restrictions. We at WIRED love meal kits, because they're all about modern convenience--you can eat what you want, even if you're on a specialty diet or have strong food preferences, without ever leaving your house. Gone are the days of grocery shopping and scouring online for recipes; these contemporary plant-based meal kit services do the heavy lifting for you using curated menus and algorithms, with choices for both premade microwavable meals and kits where you do the cooking yourself. Some plant-based meal kit services, like Hungryroot, use AI customization to curate menus based on your specific tastes. Others, like Daily Harvest, have a set selection of choices so you can always keep your freezer stocked with plant-based, gluten-free meals to have on hand. I'm vegan, so I know how difficult it can be to find new recipes that will actually taste good without breaking the bank. Plus, plant-based meal kits are a great way to try out new foods and recipes, especially if you're looking to switch to a healthier diet in the new year.

There are other approaches (e.g., [ Here, if all team members play strategies according to an NE minimizing the adversary's utility, the Eq.(1c) ensures that binary variable This space is represented by Eq.(1), which involves nonlinear terms in Eq.(1a) Section 3.4 shows that our techniques can significantly reduce the time The procedure of CRM is shown in Algorithm 2, which is illustrated in Appendix A. A collection N of subsets of players is a binary collection if: 1. { i | i N } N ; Eqs.(1b)-(1g), (3), and (4) is the space of NEs. Example 1 provides an example of N .

In this paper, we focus on computing an NE that optimizes a given objective function.

A reason is that relevance (or utilities) input to ranking algorithms may be influenced by human or societal biases, leading to output rankings that skew representations of socially-salient, and often legally-protected, groups such as women and Black people [55].

Recently, after an update that was supposed to make ChatGPT "better at guiding conversations toward productive outcomes," according to release notes from OpenAI, the bot couldn't stop telling users how brilliant their bad ideas were. ChatGPT reportedly told one person that their plan to sell literal "shit on a stick" was "not just smart--it's genius." Many more examples cropped up, and OpenAI rolled back the product in response, explaining in a blog post that "the update we removed was overly flattering or agreeable--often described as sycophantic." The company added that the chatbot's system would be refined and new guardrails would be put into place to avoid "uncomfortable, unsettling" interactions. But this was not just a ChatGPT problem. Sycophancy is a common feature of chatbots: A 2023 paper by researchers from Anthropic found that it was a "general behavior of state-of-the-art AI assistants," and that large language models sometimes sacrifice "truthfulness" to align with a user's views.

With the advent of large language models (LLMs), there has been a great deal of interest in applying them to solve difficult programming tasks. Recent work has demonstrated their potential at program optimization, a key challenge in programming languages research. We propose a blackbox adaptation method called Retrieval Augmented Search (RAS) that performs beam search over candidate optimizations; at each step, it retrieves in-context examples from a given training dataset of slow-fast program pairs to guide the LLM. Critically, we find that performing contextual retrieval based on an LLM-generated natural language description significantly outperforms retrieval based on the source code. In addition, we propose a method called AEGIS for improving interpretability by decomposing training examples into "atomic edits" that are significantly more incremental in nature. We show that RAS performs 1.8$\times$ better than prior state-of-the-art blackbox adaptation strategies, and that AEGIS performs 1.37$\times$ better while performing significantly smaller edits.

It appears everywhere you go; artificial intelligence (AI) seems to be the only two words on everyone's lips. From the rise in AI-powered chatbots to the new era of computer-generated art, it's hard to turn a blind eye to – what could be – the future of technology. However, according to a new report by Slate, AI still has a long way to go before it is considered an adequate extension of human intelligence. Slate journalist, Heather Tal Murphy, investigated AI's inability to create hands and found something even more disturbing. Long-standing rumors that AI will replace designers – ultimately making them obsolete – came to a halt after social media discovered the program's inability to create realistic hands.

An international team of researchers is applying artificial intelligence techniques to design energy-efficient district heat pump systems that better serve human needs and behaviors while reducing the carbon footprint of buildings. The $1.5-million project is funded by the National Science Foundation's (NSF) Partnerships for International Research and Education (PIRE) program and led by Zheng O'Neill of the J. Mike Walker '66 Department of Mechanical Engineering at Texas A&M University. The PIRE program funds only an estimated 10-15 projects nationwide at a time, according to NSF. The research is also supported by the Texas A&M Engineering Experiment Station's Energy Systems Laboratory, of which O'Neill is an associate director. The project focuses on the technology of district heat pump systems, which distribute energy to buildings through a system of heat pumps and insulated networked pipes.

The fair-ranking problem, which asks to rank a given set of items to maximize utility subject to group fairness constraints, has received attention in the fairness, information retrieval, and machine learning literature. Recent works, however, observe that errors in socially-salient (including protected) attributes of items can significantly undermine fairness guarantees of existing fair-ranking algorithms and raise the problem of mitigating the effect of such errors. We study the fair-ranking problem under a model where socially-salient attributes of items are randomly and independently perturbed. We present a fair-ranking framework that incorporates group fairness requirements along with probabilistic information about perturbations in socially-salient attributes. We provide provable guarantees on the fairness and utility attainable by our framework and show that it is information-theoretically impossible to significantly beat these guarantees. Our framework works for multiple non-disjoint attributes and a general class of fairness constraints that includes proportional and equal representation. Empirically, we observe that, compared to baselines, our algorithm outputs rankings with higher fairness, and has a similar or better fairness-utility trade-off compared to baselines.