Parameter regularization or allocation methods are effective in overcoming catastrophic forgetting in lifelong learning. However, they solve all tasks in a sequence uniformly and ignore the differences in the learning difficulty of different tasks. So parameter regularization methods face significant forgetting when learning a new task very different from learned tasks, and parameter allocation methods face unnecessary parameter overhead when learning simple tasks. In this paper, we propose the Parameter Allocation & Regularization (PAR), which adaptively select an appropriate strategy for each task from parameter allocation and regularization based on its learning difficulty. A task is easy for a model that has learned tasks related to it and vice versa. We propose a divergence estimation method based on the Nearest-Prototype distance to measure the task relatedness using only features of the new task. Moreover, we propose a time-efficient relatedness-aware sampling-based architecture search strategy to reduce the parameter overhead for allocation. Experimental results on multiple benchmarks demonstrate that, compared with SOTAs, our method is scalable and significantly reduces the model's redundancy while improving the model's performance. Further qualitative analysis indicates that PAR obtains reasonable task-relatedness.

We consider online reinforcement learning in Mean-Field Games (MFGs). Unlike traditional approaches, we alleviate the need for a mean-field oracle by developing an algorithm that approximates the Mean-Field Equilibrium (MFE) using the single sample path of the generic agent. We call this {\it Sandbox Learning}, as it can be used as a warm-start for any agent learning in a multi-agent non-cooperative setting. We adopt a two time-scale approach in which an online fixed-point recursion for the mean-field operates on a slower time-scale, in tandem with a control policy update on a faster time-scale for the generic agent. Given that the underlying Markov Decision Process (MDP) of the agent is communicating, we provide finite sample convergence guarantees in terms of convergence of the mean-field and control policy to the mean-field equilibrium. The sample complexity of the Sandbox learning algorithm is $\tilde{\mathcal{O}}(\epsilon^{-4})$ where $\epsilon$ is the MFE approximation error. This is similar to works which assume access to oracle. Finally, we empirically demonstrate the effectiveness of the sandbox learning algorithm in diverse scenarios, including those where the MDP does not necessarily have a single communicating class.

How to behave efficiently and flexibly is a central problem for understanding biological agents and creating intelligent embodied AI. It has been well known that behavior can be classified as two types: reward-maximizing habitual behavior, which is fast while inflexible; and goal-directed behavior, which is flexible while slow. Conventionally, habitual and goal-directed behaviors are considered handled by two distinct systems in the brain. Here, we propose to bridge the gap between the two behaviors, drawing on the principles of variational Bayesian theory. We incorporate both behaviors in one framework by introducing a Bayesian latent variable called "intention". The habitual behavior is generated by using prior distribution of intention, which is goal-less; and the goal-directed behavior is generated by the posterior distribution of intention, which is conditioned on the goal. Building on this idea, we present a novel Bayesian framework for modeling behaviors. Our proposed framework enables skill sharing between the two kinds of behaviors, and by leveraging the idea of predictive coding, it enables an agent to seamlessly generalize from habitual to goal-directed behavior without requiring additional training. The proposed framework suggests a fresh perspective for cognitive science and embodied AI, highlighting the potential for greater integration between habitual and goal-directed behaviors.

Online federated learning (FL) enables geographically distributed devices to learn a global shared model from locally available streaming data. Most online FL literature considers a best-case scenario regarding the participating clients and the communication channels. However, these assumptions are often not met in real-world applications. Asynchronous settings can reflect a more realistic environment, such as heterogeneous client participation due to available computational power and battery constraints, as well as delays caused by communication channels or straggler devices. Further, in most applications, energy efficiency must be taken into consideration. Using the principles of partial-sharing-based communications, we propose a communication-efficient asynchronous online federated learning (PAO-Fed) strategy. By reducing the communication overhead of the participants, the proposed method renders participation in the learning task more accessible and efficient. In addition, the proposed aggregation mechanism accounts for random participation, handles delayed updates and mitigates their effect on accuracy. We prove the first and second-order convergence of the proposed PAO-Fed method and obtain an expression for its steady-state mean square deviation. Finally, we conduct comprehensive simulations to study the performance of the proposed method on both synthetic and real-life datasets. The simulations reveal that in asynchronous settings, the proposed PAO-Fed is able to achieve the same convergence properties as that of the online federated stochastic gradient while reducing the communication overhead by 98 percent.

This course offers a brief introduction to the multivariate calculus required to build many common machine learning techniques. We start at the very beginning with a refresher on the "rise over run" formulation of a slope, before converting this to the formal definition of the gradient of a function. We then start to build up a set of tools for making calculus easier and faster. Next, we learn how to calculate vectors that point up hill on multidimensional surfaces and even put this into action using an interactive game. We take a look at how we can use calculus to build approximations to functions, as well as helping us to quantify how accurate we should expect those approximations to be.

Given the new oracle that is ChatGPT, you may often find yourself tasked with creating prompts for various applications. One of the most significant challenges in this regard is crafting prompts that effectively communicate your requirements and elicit the desired response. In this article, I will provide a comprehensive guide on how to write high-quality prompts for software development, specifically for the ChatGPT language model. Our aim is to help you improve your skills as a prompt engineer, moving beyond generic advice and offering practical tips and examples. To create effective prompts, it is essential to understand the AI language model you are working with.

Data image augmentation is a technique used in computer vision and deep learning to increase the amount and diversity of data available for training a model. This paper presents an overview of data image augmentation and provides a tutorial on how to perform data image augmentation in Python using the Keras.preprocessing.image The paper also includes a discussion on the benefits and limitations of data image augmentation and provides tips on how to use it effectively. In recent years, computer vision and deep learning have made significant strides in accurately classifying and detecting objects in images. One of the key factors that contribute to the success of these techniques is the availability of large and diverse datasets for training models.

We study the effectiveness of neural sequence models for premise selection in automated theorem proving, one of the main bottlenecks in the formalization of mathematics. We propose a two stage approach for this task that yields good results for the premise selection task on the Mizar corpus while avoiding the handengineered features of existing state-of-the-art models. To our knowledge, this is the first time deep learning has been applied to theorem proving on a large scale.

The mean and variance of the series are constant over time. The series has a constant trend over time. The auto-covariance function of the series is dependent on time. The series has a periodic pattern over time. A moving average uses past errors, while an autoregressive model uses past values of the dependent variable. A moving average uses only one past value, while an autoregressive model uses multiple past values.

Today I am happy to announce that a new tidymodels-centric version of my free, online, interactive course, Supervised Machine Learning: Case Studies in R, has been published! This is at least the third version of this course I've built at this point but I believe it to be the best, in terms of how it communicates machine learning concepts and how useful to your real-world problems the demonstrated code will be. Similar to the last time I launched this course, it provides four case studies using data from the real world for you to practice your predictive modeling skills. One question we sometimes field from R users is about choosing to use tidymodels vs. caret. The original version of my course mostly used caret, and caret is a stable and broadly used framework for modeling and machine learning in R.