In this project, our goal is to determine how to leverage the world-knowledge of pretrained large language models for efficient and robust learning in multiagent decision making. We examine this in a taxi routing and assignment problem where agents must decide how to best pick up passengers in order to minimize overall waiting time. While this problem is situated on a graphical road network, we show that with the proper prompting zero-shot performance is quite strong on this task. Furthermore, with limited fine-tuning along with the one-at-a-time rollout algorithm for look ahead, LLMs can out-compete existing approaches with 50 times fewer environmental interactions. We also explore the benefits of various linguistic prompting approaches and show that including certain easy-to-compute information in the prompt significantly improves performance. Finally, we highlight the LLM's built-in semantic understanding, showing its ability to adapt to environmental factors through simple prompts.

The proper planning of different types of public transportation such as metro, highway, waterways, and so on, can increase the efficiency, reduce the congestion and improve the safety of the country. There are certain challenges associated with route planning, such as high cost of implementation, need for adequate resource & infrastructure and resistance to change. The goal of this research is to examine the working, applications, complexity factors, advantages & disadvantages of Floyd- Warshall, Bellman-Ford, Johnson, Ant Colony Optimization (ACO), Particle Swarm Optimization (PSO), & Grey Wolf Optimizer (GWO), to find the best choice for the above application. In this paper, comparative analysis of above-mentioned algorithms is presented. The Floyd-Warshall method and ACO algorithm are chosen based on the comparisons. Also, a combination of modified Floyd-Warshall with ACO algorithm is proposed. The proposed algorithm showed better results with less time complexity, when applied on randomly structured points within a boundary called quasi-structured points. In addition, this paper also discusses the future works of integrating Floyd-Warshall with ACO to develop a real-time model for overcoming above mentioned-challenges during transportation route planning.

In order to define an optimization problem, you need three things: variables, constraints and an objective. The variables can take different values, the solver will try to find the best values for the variables. Constraints are things that are not allowed or boundaries, by setting these correctly you are sure that you will find a solution you can actually use in real life. The objective is the goal you have in the optimization problem, this is what you want to maximize or minimize. If it's not completely clear by now, here is a more thorough introduction.

When it comes to getting from A to B on foot, it turns out we're wired not to take the shortest route but rather the'pointiest path'. The reason for this, researchers say, is because human brains are not optimised to navigate cities. Instead, pedestrians appear to choose paths that seem to point most directly toward their destination, even if those routes end up being longer. Researchers at MIT, who based their study on a dataset of more than 14,000 people going about their daily lives, called this the'pointiest path'. An MIT study suggests our brains are not optimized to calculate the shortest possible route when navigating on foot.

Recent advancements in quantum computing have driven the scientific community's quest to solve a certain class of complex problems for which quantum computers would be better suited than traditional supercomputers. To improve the efficiency with which quantum computers can solve these problems, scientists are investigating the use of artificial intelligence approaches. In a new study, scientists at the U.S. Department of Energy's (DOE) Argonne National Laboratory have developed a new algorithm based on reinforcement learning to find the optimal parameters for the Quantum Approximate Optimization Algorithm (QAOA), which allows a quantum computer to solve certain combinatorial problems such as those that arise in materials design, chemistry and wireless communications. "Combinatorial optimization problems are those for which the solution space gets exponentially larger as you expand the number of decision variables," said Argonne computer scientist Prasanna Balaprakash. "In one traditional example, you can find the shortest route for a salesman who needs to visit a few cities once by enumerating all possible routes, but given a couple thousand cities, the number of possible routes far exceeds the number of stars in the universe; even the fastest supercomputers cannot find the shortest route in a reasonable time."

Recent advancements in quantum computing have driven the scientific community's quest to solve a certain class of complex problems for which quantum computers would be better suited than traditional supercomputers. To improve the efficiency with which quantum computers can solve these problems, scientists are investigating the use of artificial intelligence approaches. In a new study, scientists at the U.S. Department of Energy's (DOE) Argonne National Laboratory have developed a new algorithm based on reinforcement learning to find the optimal parameters for the Quantum Approximate Optimization Algorithm (QAOA), which allows a quantum computer to solve certain combinatorial problems such as those that arise in materials design, chemistry and wireless communications. "It's a bit like having a self-driving car in traffic; the algorithm can detect when it needs to make adjustments in the'dials' it uses to do the computation." "Combinatorial optimization problems are those for which the solution space gets exponentially larger as you expand the number of decision variables," said Argonne computer scientist Prasanna Balaprakash.

It is one of the most simple organisms on Earth. But the amoeba, a single-celled organism consisting mostly of gelatinous protoplasm, could be far smarter that thought - and change computing forever. Researchers found they have unique computing abilities that could one day rival conventional computers. The researchers adapted the problem so the amoeba, which can deform its body, was able to use a specially designed chip with 64 'legs', pictured Amoeba, are a single-celled organism consisting mostly of gelatinous protoplasm. The particular type of amoeba that the scientists used was a plasmodium or'true slime mold,' which weighs about 12 mg and consumes oat flakes.

One of the longest-running challenges in the logistics industry is finding the shortest routes. First articulated in the 1930s, the "traveling salesman problem" seeks to deduce the shortest route connecting a group of cities to ensure optimal use of time and resources. Karim Beguir, co-founder and CEO of London-based AI startup InstaDeep, told GPU Technology Conference attendees this week that GPU-powered deep learning and reinforcement learning may have the answer. Previous efforts to address the traveling salesman problem include optimization solvers, heuristics and Monte Carlo Tree Search algorithms. But, according to Beguir, these approaches all suffer from the same shortcoming: They don't learn.

The world of AI has a lot of things around it to thank for its existence in our technological landscape of today. Not only have humans spent decades of research perfecting the mathematical calculations to make these wonderfully complex learning algorithms work but during this time we have looked further than our own species as inspiration to make the next generation of intelligent presence on our planet. Mother Nature, and all that it encompasses, has it's roots firmly planted in the workings of Artificial Intelligence -- and it's here to stay. They go into incredible, high definition detail about the behaviours and properties of the Earth's many inhabitants, and they allow us to understand how they fit into the natural ecosystem and work together in order to allow our planet to flourish -- to make it Earth. Now I'm no Sir David Attenborough, but I'm still going to take you on a wildlife documentary of my own.

A widely studied non-deterministic polynomial time (NP) hard problem lies in finding a route between the two nodes of a graph. Often meta-heuristics algorithms such as $A^{*}$ are employed on graphs with a large number of nodes. Here, we propose a deep recurrent neural network architecture based on the Sequence-2-Sequence (Seq2Seq) model, widely used, for instance in text translation. Particularly, we illustrate that utilising a context vector that has been learned from two different recurrent networks enables increased accuracies in learning the shortest route of a graph. Additionally, we show that one can boost the performance of the Seq2Seq network by smoothing the loss function using a homotopy continuation of the decoder's loss function.