Cathy Wu is the Gilbert W. Winslow Assistant Professor of Civil and Environmental Engineering and a member of the MIT Institute for Data, Systems, and Society. Cathy Wu is the Gilbert W. Winslow Assistant Professor of Civil and Environmental Engineering and a member of the MIT Institute for Data, Systems, and Society. As an undergraduate, Wu won MIT's toughest robotics competition, and as a graduate student took the University of California at Berkeley's first-ever course on deep reinforcement learning. Now back at MIT, she's working to improve the flow of robots in Amazon warehouses under the Science Hub, a new collaboration between the tech giant and the MIT Schwarzman College of Computing. Outside of the lab and classroom, Wu can be found running, drawing, pouring lattes at home, and watching YouTube videos on math and infrastructure via 3Blue1Brown and Practical Engineering.

Do your non-player characters lack drive and ambition? Are they slow, stupid and constantly banging their heads against the wall? Then this course is for you. Join Penny as she explains, demonstrates and assists you in creating your very own NPCs in Unity with C#. All you need is a sound knowledge of Unity, C# and the ability to add two numbers together. In this course, Penny reveals the most popular AI techniques used for creating believable character behaviour in games using her internationally acclaimed teaching style and knowledge from over 25 years working with games, graphics and having written two award winning books on games AI.

I will recommend this class to any one looking towards Data Science" "This course so far is breaking down the content into smart bite-size pieces and the professor explains everything patiently and gives just enough background so that I do not feel lost." "This course is really good for me. it is easy to understand and it covers a wide range of NLP topics from the basics, machine learning to Deep Learning. The codes used is practical and useful. I definitely satisfy with the content and surely recommend to everyone who is interested in Natural Language Processing"

Welcome to Deep Reinforcement Learning 2.0! In this course, we will learn and implement a new incredibly smart AI model, called the Twin-Delayed DDPG, which combines state of the art techniques in Artificial Intelligence including continuous Double Deep Q-Learning, Policy Gradient, and Actor Critic. The model is so strong that for the first time in our courses, we are able to solve the most challenging virtual AI applications (training an ant/spider and a half humanoid to walk and run across a field). In this part we will study all the fundamentals of Artificial Intelligence which will allow you to understand and master the AI of this course. These include Q-Learning, Deep Q-Learning, Policy Gradient, Actor-Critic and more.

We extend and combine several tools of the literature to design fast, adaptive, anytime and scale-free online learning algorithms. Scale-free regret bounds must scale linearly with the maximum loss, both toward large losses and toward very small losses. Adaptive regret bounds demonstrate that an algorithm can take advantage of easy data and potentially have constant regret. We seek to develop fast algorithms that depend on as few parameters as possible, in particular they should be anytime and thus not depend on the time horizon. Our first and main tool, isotuning, is a generalization of the idea of balancing the trade-off of the regret. We develop a set of tools to design and analyze such learning rates easily and show that they adapts automatically to the rate of the regret (whether constant, $O(\log T)$, $O(\sqrt{T})$, etc.) within a factor 2 of the optimal learning rate in hindsight for the same observed quantities. The second tool is an online correction, which allows us to obtain centered bounds for many algorithms, to prevent the regret bounds from being vacuous when the domain is overly large or only partially constrained. The last tool, null updates, prevents the algorithm from performing overly large updates, which could result in unbounded regret, or even invalid updates. We develop a general theory using these tools and apply it to several standard algorithms. In particular, we (almost entirely) restore the adaptivity to small losses of FTRL for unbounded domains, design and prove scale-free adaptive guarantees for a variant of Mirror Descent (at least when the Bregman divergence is convex in its second argument), extend Adapt-ML-Prod to scale-free guarantees, and provide several other minor contributions about Prod, AdaHedge, BOA and Soft-Bayes.

Motivated by the problem of online canonical correlation analysis, we propose the \emph{Stochastic Scaled-Gradient Descent} (SSGD) algorithm for minimizing the expectation of a stochastic function over a generic Riemannian manifold. SSGD generalizes the idea of projected stochastic gradient descent and allows the use of scaled stochastic gradients instead of stochastic gradients. In the special case of a spherical constraint, which arises in generalized eigenvector problems, we establish a nonasymptotic finite-sample bound of $\sqrt{1/T}$, and show that this rate is minimax optimal, up to a polylogarithmic factor of relevant parameters. On the asymptotic side, a novel trajectory-averaging argument allows us to achieve local asymptotic normality with a rate that matches that of Ruppert-Polyak-Juditsky averaging. We bring these ideas together in an application to online canonical correlation analysis, deriving, for the first time in the literature, an optimal one-time-scale algorithm with an explicit rate of local asymptotic convergence to normality. Numerical studies of canonical correlation analysis are also provided for synthetic data.

We study universal consistency of non-i.i.d. processes in the context of online learning. A stochastic process is said to admit universal consistency if there exists a learner that achieves vanishing average loss for any measurable response function on this process. When the loss function is unbounded, Blanchard et al. showed that the only processes admitting strong universal consistency are those taking a finite number of values almost surely. However, when the loss function is bounded, the class of processes admitting strong universal consistency is much richer and its characterization could be dependent on the response setting (Hanneke). In this paper, we show that this class of processes is independent from the response setting thereby closing an open question (Hanneke, Open Problem 3). Specifically, we show that the class of processes that admit universal online learning is the same for binary classification as for multiclass classification with countable number of classes. Consequently, any output setting with bounded loss can be reduced to binary classification. Our reduction is constructive and practical. Indeed, we show that the nearest neighbor algorithm is transported by our construction. For binary classification on a process admitting strong universal learning, we prove that nearest neighbor successfully learns at least all finite unions of intervals.

A challenge for many young adults is to find the right institution to follow higher education. Global university rankings are commonly used, but inefficient tool, for they do not consider a person's preferences and needs. For example, some persons pursue prestige in their higher education, while others prefer proximity. This paper develops and evaluates a university recommender system, eliciting user preferences as ratings to build predictive models and to generate personalized university ranking lists. In Study 1, we performed offline evaluation on a rating dataset to determine which recommender approaches had the highest predictive value. In Study 2, we selected three algorithms to produce different university recommendation lists in our online tool, asking our users to compare and evaluate them in terms of different metrics (Accuracy, Diversity, Perceived Personalization, Satisfaction and Novelty). We show that a SVD algorithm scores high on accuracy and perceived personalization, while a KNN algorithm scores better on novelty. We also report findings on preferred university features.

A state machine model is a mathematical model that groups all possible system occurrences, called states. The course emphasizes project-based learning, learning by doing. The goal of this course is to introduce an event-driven programming paradigm using simple and hierarchical state machines. After going through this course, you will be trained to apply the state machine approach to solve your complex embedded systems projects. If you are a beginner in the field of embedded systems, then you can take our courses in the below-mentioned order.

Create your first programs easily in the latest version of Python. The Python course covers everything you need to know as a beginner in Python and in programming itself to successfully write your first programs and codes. The focus is on practice so that you can primarily use the functions of Python in real life to solve your problems and write your own code, in addition to theoretical lections. We'll focus on all the important elements, functions, and operators that a beginner should know in order to code successfully in Python. We also look at the foundations for further programming in areas such as machine learning, data science, and Django.