For example, one game we have experimented on is a puzzle game called Refraction, Reinforcement Learning is beginning to be applied which teaches kids how to multiply fractions by splitting laser outside traditional domains such as robotics, and beams. In this case, we need to choose right sequence of puzzles into human-centric domains such as healthcare and to give to students such that they complete the most concepts education. In these domains, two problems are critical successfully. Making this decision is difficult because to address: We must be able to evaluate algorithms of the large amount of information we can collect about each with a collection of prior data if one is available, student, most of which is not very useful for the task at hand. and we must devise algorithms that carefully Refraction and our other educational games, such as Treefrog trade off exploration and exploitation in such a way Treasure, have each been played by hundreds of thousands that they are guaranteed to converge to optimal behavior of players, providing ample opportunity for learning how to quickly, while retaining very good performance improve these games using data.

In partial label learning, each training example is associated with a set of candidate labels, among which only one is valid. An intuitive strategy to learn from partial label examples is to treat all candidate labels equally and make prediction by averaging their modeling outputs. Nonetheless, this strategy may suffer from the problem that the modeling output from the valid label is overwhelmed by those from the false positive labels. In this paper, an instance-based approach named IPAL is proposed by directly disambiguating the candidate label set. Briefly, IPAL tries to identify the valid label of each partial label example via an iterative label propagation procedure, and then classifies the unseen instance based on minimum error reconstruction from its nearest neighbors. Extensive experiments show that IPAL compares favorably against the existing instance-based as well as other state-of-the-art partial label learning approaches.

In this paper we investigate the use of coactive learning in a multitask setting. In coactive learning, an expert presents the learner with a problem and the learner returns a candidate solution. The expert then improves on the solution if necessary and presents the improved solution to the learner. The goal for the learner is to learn to produce solutions which cannot be further improved by the expert while minimizing the average expert effort. In this paper, we consider the setting where there are multiple experts (tasks), and in each iteration one expert presents a problem to the learner. While the experts are expected to have different solution preferences, they are also assumed to share similarities, which should enable generalization across experts. We analyze several algorithms for this setting and derive bounds on the average expert effort during learning. Our main contribution is the balanced Perceptron algorithm, which is the first coactive learning algorithm that is both able to generalize across experts when possible, while also guaranteeing convergence to optimal solutions for individual experts. Our experiments in three domains confirm that this algorithm is effective in the multitask setting, compared to natural baselines.

Raven’s Progressive Matrices (RPMs) are a popular family of general intelligence tests, and provide a non-verbal measure of a test subject’s reasoning abilities. Traditionally RPMs have been manually designed. To make them readily available for both practice and examination, we tackle the problem of automatically synthesizing RPMs. Our goal is to efficiently generate a large number of RPMs that are authentic (i.e. similar to manually written problems), interesting (i.e. diverse in terms of difficulty), and well-formed (i.e unambiguous). The main technical challenges are: How to formalize RPMs to accommodate their seemingly enormous diversity, and how to define and enforce their validity? To this end, we (1) introduce an abstract representation of RPMs using first-order logic, and (2) restrict instantiations to only valid RPMs. We have realized our approach and evaluated its efficiency and effectiveness. We show that our system can generate hundreds of valid problems per second with varying levels of difficulty. More importantly, we show, via a user study with 24 participants, that the generated problems are statistically indistinguishable from actual problems. This work is an exciting instance of how logic and reasoning may aid general learning.

The integration of the various specialized components of cognitive systems poses a challenge, in particular for those architectures that combine planning, inference, and human-computer interaction (HCI). An approach is presented that exploits a single source of common knowledge contained in an ontology. Based upon the knowledge contained in it, specialized domain models for the cognitive systems' components can be generated automatically. Our integration targets planning in the form of hierarchical planning, being well-suited for HCI as it mimics planning done by humans. We show how the hierarchical structures of such planning domains can be (partially) inferred from declarative background knowledge. The same ontology furnishes the structure of the interaction between the cognitive system and the user. First, explanations of plans presented to users are enhanced by ontology explanations. Second, a dialog domain is created from the ontology coherent with the planning domain. We demonstrate the application of our technique in a fitness training scenario.

A major challenge in Content-Based Image Retrieval (CBIR) is to bridge the semantic gap between low-level image contents and high-level semantic concepts. Although researchers have investigated a variety of retrieval techniques using different types of features and distance functions, no single best retrieval solution can fully tackle this challenge. In a real-world CBIR task, it is often highly desired to combine multiple types of different feature representations and diverse distance measures in order to close the semantic gap. In this paper, we investigate a new framework of learning to rank for CBIR, which aims to seek the optimal combination of different retrieval schemes by learning from large-scale training data in CBIR. We first formulate the problem formally as a learning to rank task, which can be solved in general by applying the existing batch learning to rank algorithms from text information retrieval (IR). To further address the scalability towards large-scale online CBIR applications, we present a family of online learning to rank algorithms, which are significantly more efficient and scalable than classical batch algorithms for large-scale online CBIR. Finally, we conduct an extensive set of experiments, in which encouraging results show that our technique is effective, scalable and promising for large-scale CBIR.

In Massively Open Online Courses (MOOCs) TA resources are limited; most MOOCs use peer assessments to grade assignments. Students have to divide up their time between working on their own homework and grading others. If there is no risk of being caught and penalized, students have no reason to spend any time grading others Course staff want to incentivize students to balance their time between course work and peer grading. They may do so by auditing students, ensuring that they perform grading correctly. One would not want students to invest too much time on peer grading, as this would result in poor course performance. We present the first model of strategic auditing in peer grading, modeling the student's choice of effort in response to a grader's audit levels as a Stackelberg game with multiple followers. We demonstrate that computing the equilibrium for this game is computationally hard. We then provide a PTAS in order to compute an approximate solution to the problem of allocating audit levels. However, we show that this allocation does not necessarily maximize social welfare; in fact, there exist settings where course auditor utility is arbitrarily far from optimal under an approximately optimal allocation. To circumvent this issue, we present a natural condition that guarantees that approximately optimal TA allocations guarantee approximately optimal welfare for the course auditors.

Reinforcement learning describes how a learning agent can achieve optimal behaviour based on interactions with its environment and reward feedback. A limiting factor in reinforcement learning as employed in artificial intelligence is the need for an often prohibitively large number of environment samples before the agent reaches a desirable level of performance. Learning from demonstration is an approach that provides the agent with demonstrations by a supposed expert, from which it should derive suitable behaviour. Yet, one of the challenges of learning from demonstration is that no guarantees can be provided for the quality of the demonstrations, and thus the learned behavior. In this paper, we investigate the intersection of these two approaches, leveraging the theoretical guarantees provided by reinforcement learning, and using expert demonstrations to speed up this learning by biasing exploration through a process called reward shaping. This approach allows us to leverage human input without making an erroneous assumption regarding demonstration optimality. We show experimentally that this approach requires significantly fewer demonstrations, is more robust against suboptimality of demonstrations, and achieves much faster learning than the recently developed HAT algorithm.

Goal Recognition is the task of inferring an actor's goals given some or all of the actor's observed actions. There is considerable interest in Goal Recognition for use in intelligent personal assistants, smart environments, intelligent tutoring systems, and monitoring user's needs. In much of this work, the actor's observed actions are compared against a generated library of plans. Recent work by Ramirez and Geffner makes use of AI planning to determine how closely a sequence of observed actions matches plans for each possible goal. For each goal, this is done by comparing the cost of a plan for that goal with the cost of a plan for that goal that includes the observed actions. This approach yields useful rankings, but is impractical for real-time goal recognition in large domains because of the computational expense of constructing plans for each possible goal. In this paper, we introduce an approach that propagates cost and interaction information in a plan graph, and uses this information to estimate goal probabilities. We show that this approach is much faster, but still yields high quality results.

Geometry reasoning and proof form a major and challenging component in the K-121 mathematics curriculum. Although several computerized systems exist that help students learn and practice general geometry concepts, they do not target geometry proof problems, which are more advanced and difficult. Powerful geometry theorem provers also exist, however they typically employ advanced algebraic methods and generate complex, difficult to understand proofs, and thus do not meet general K-12 students’ educational needs. This paper tackles these weaknesses of prior systems by introducing a geometry proof system, iGeoTutor, capable of generating human-readable elementary proofs, i.e. proofs using standard Euclidean axioms. We have gathered 77 problems in total from various sources, including ones unsolvable by other systems and from Math competitions. iGeoTutor solves all but two problems in under two minutes each, and more importantly, demonstrates a much more effective and intelligent proof search than prior systems. We have also conducted a pilot study with 12 high school students, and the results show that iGeoTutor provides a clear benefit in helping students learn geometry proofs. We are in active discussions with Khan Academy and local high schools for possible adoption of iGeo-Tutor in real learning environments.