Autonomous agents situated in real-world environments must be able to master large repertoires of skills. While a single short skill can be learned quickly, it would be impractical to learn every task independently. Instead, the agent should share knowledge across behaviors such that each task can be learned efficiently, and such that the resulting model can generalize to new tasks, especially ones that are compositions or subsets of tasks seen previously. A policy conditioned on a goal or demonstration has the potential to share knowledge between tasks if it sees enough diversity of inputs. However, these methods may not generalize to a more complex task at test time. We introduce compositional plan vectors (CPVs) to enable a policy to perform compositions of tasks without additional supervision. CPVs represent trajectories as the sum of the subtasks within them. We show that CPVs can be learned within a one-shot imitation learning framework without any additional supervision or information about task hierarchy, and enable a demonstration-conditioned policy to generalize to tasks that sequence twice as many skills as the tasks seen during training. Analogously to embeddings such as word2vec in NLP, CPVs can also support simple arithmetic operations -- for example, we can add the CPVs for two different tasks to command an agent to compose both tasks, without any additional training.

Language is an important tool for humans to share knowledge. We propose a meta-learning method that shares knowledge across supervised learning tasks using feature descriptions written in natural language, which have not been used in the existing meta-learning methods. The proposed method improves the predictive performance on unseen tasks with a limited number of labeled data by meta-learning from various tasks. With the feature descriptions, we can find relationships across tasks even when their feature spaces are different. The feature descriptions are encoded using a language model pretrained with a large corpus, which enables us to incorporate human knowledge stored in the corpus into meta-learning. In our experiments, we demonstrate that the proposed method achieves better predictive performance than the existing meta-learning methods using a wide variety of real-world datasets provided by the statistical office of the EU and Japan.

Language is an important tool for humans to share knowledge. We propose a meta-learning method that shares knowledge across supervised learning tasks using feature descriptions written in natural language, which have not been used in the existing meta-learning methods. The proposed method improves the predictive performance on unseen tasks with a limited number of labeled data by meta-learning from various tasks. With the feature descriptions, we can find relationships across tasks even when their feature spaces are different. The feature descriptions are encoded using a language model pretrained with a large corpus, which enables us to incorporate human knowledge stored in the corpus into meta-learning.

Autonomous agents situated in real-world environments must be able to master large repertoires of skills. While a single short skill can be learned quickly, it would be impractical to learn every task independently. Instead, the agent should share knowledge across behaviors such that each task can be learned efficiently, and such that the resulting model can generalize to new tasks, especially ones that are compositions or subsets of tasks seen previously. A policy conditioned on a goal or demonstration has the potential to share knowledge between tasks if it sees enough diversity of inputs. However, these methods may not generalize to a more complex task at test time. We introduce compositional plan vectors (CPVs) to enable a policy to perform compositions of tasks without additional supervision.

The EU's proposed Artificial Intelligence Act plans to restrict open-source AI. The proposed – and still debated – Artificial Intelligence Act (AIA) from the EU touches upon the regulation of open-source AI. But enforcing strict restrictions on the sharing and distribution of open-source general-purpose AI (GPAI) is a completely retrograde step. It is like rewinding the world back by 30 years. Open-source culture is the only reason why mankind was able to progress technology at such a light speed. Only recently AI researchers were able to embrace sharing their code for more transparency and verification but putting constraints on this movement will damage the cultural progress the scientific community has made.

There is a whole lot of talk nowadays about power, who has it, who wants it, who should have it. What you do not hear a lot about is what exactly power is and where it really comes from. We believe that everybody in a free and capitalistic democracy should have it -- and we believe that one of the surest sources of it is knowledge. Those who possess crucial information and the knowhow to put that information to use effectively will be heads and shoulders above those who do not -- and not because they are better or more deserving but simply because they have that knowledge. The fact is that knowledge is and always has been the ultimate equalizer or advantage if you had it and others did not, and the ultimate granter of access to all that you could be. It's why monks slaved by candlelight to find and preserve the words of wisdom from the Greek and Roman empires almost lost forever during the Dark Ages by rewriting every word by hand. It's why the invention of the printing press, radio and TV and, ultimately, the internet on which you are reading this article changed the world. Conversely, it's the reason why for centuries those who wanted to keep people down and out of power have tried to keep knowledge out of the hands of those they wanted to oppress. In the United States, for instance, during the era of slavery, slave codes actually made it illegal to teach slaves to read or write. And in current day Afghanistan, women are not even allowed to go to school.

Please feel free to look at other tips and news on my website, I share weekly blogs and share knowledge and wisdom on it and will be putting some Github projects to share knowledge and improve data knowledge for all people.

Recently, deep reinforcement learning (RL) algorithms have made great progress in multi-agent domain. However, due to characteristics of RL, training for complex tasks would be resource-intensive and time-consuming. To meet this challenge, mutual learning strategy between homogeneous agents is essential, which is under-explored in previous studies, because most existing methods do not consider to use the knowledge of agent models. In this paper, we present an adaptation method of the majority of multi-agent reinforcement learning (MARL) algorithms called "KnowSR" which takes advantage of the differences in learning between agents. We employ the idea of knowledge distillation (KD) to share knowledge among agents to shorten the training phase. To empirically demonstrate the robustness and effectiveness of KnowSR, we performed extensive experiments on state-of-the-art MARL algorithms in collaborative and competitive scenarios. The results demonstrate that KnowSR outperforms recently reported methodologies, emphasizing the importance of the proposed knowledge sharing for MARL.

Ride-hailing platforms generally provide various service options to customers, such as solo ride services, shared ride services, etc. It is generally expected that demands for different service modes are correlated, and the prediction of demand for one service mode can benefit from historical observations of demands for other service modes. Moreover, an accurate joint prediction of demands for multiple service modes can help the platforms better allocate and dispatch vehicle resources. Although there is a large stream of literature on ride-hailing demand predictions for one specific service mode, little efforts have been paid towards joint predictions of ride-hailing demands for multiple service modes. To address this issue, we propose a deep multi-task multi-graph learning approach, which combines two components: (1) multiple multi-graph convolutional (MGC) networks for predicting demands for different service modes, and (2) multi-task learning modules that enable knowledge sharing across multiple MGC networks. More specifically, two multi-task learning structures are established. The first one is the regularized cross-task learning, which builds cross-task connections among the inputs and outputs of multiple MGC networks. The second one is the multi-linear relationship learning, which imposes a prior tensor normal distribution on the weights of various MGC networks. Although there are no concrete bridges between different MGC networks, the weights of these networks are constrained by each other and subject to a common prior distribution. Evaluated with the for-hire-vehicle datasets in Manhattan, we show that our propose approach outperforms the benchmark algorithms in prediction accuracy for different ride-hailing modes.

Artificial intelligence has often been depicted as the Fourth Industrial Revolution. It has become a top need for organizations, with numerous effectively contributing and exploring the abilities of AI. Artificial intelligence has been developing at a fantastic rate in the course of a recent couple of years and as indicated by PwC, A.I. could signify $15.7 trillion to the worldwide economy by 2030. For a variety of reasons, for example, government subsidizing, talent location and research opportunities, a few urban areas have become pioneers in the development of this rapidly growing technology. As an AI expert, one should always be updated with the latest. Having said that, there are online groups and communities where AI enthusiasts can share knowledge, ideas, problems, insights etc.