Deep reinforcement learning (RL) algorithms can use high-capacity deep networks to learn directly from image observations. However, these kinds of observation spaces present a number of challenges in practice, since the policy must now solve two problems: a representation learning problem, and a task learning problem. In this paper, we aim to explicitly learn representations that can accelerate reinforcement learning from images. We propose the stochastic latent actor-critic (SLAC) algorithm: a sample-efficient and high-performing RL algorithm for learning policies for complex continuous control tasks directly from high-dimensional image inputs. SLAC learns a compact latent representation space using a stochastic sequential latent variable model, and then learns a critic model within this latent space. By learning a critic within a compact state space, SLAC can learn much more efficiently than standard RL methods. The proposed model improves performance substantially over alternative representations as well, such as variational autoencoders. In fact, our experimental evaluation demonstrates that the sample efficiency of our resulting method is comparable to that of model-based RL methods that directly use a similar type of model for control. Furthermore, our method outperforms both model-free and model-based alternatives in terms of final performance and sample efficiency, on a range of difficult image-based control tasks.

Automatic question generation according to an answer within the given passage is useful for many applications, such as question answering system, dialogue system, etc. Current neural-based methods mostly take two steps which extract several important sentences based on the candidate answer through manual rules or supervised neural networks and then use an encoder-decoder framework to generate questions about these sentences. These approaches neglect the semantic relations between the answer and the context of the whole passage which is sometimes necessary for answering the question. To address this problem, we propose the Weak Supervision Enhanced Generative Network (WeGen) which automatically discovers relevant features of the passage given the answer span in a weakly supervised manner to improve the quality of generated questions. More specifically, we devise a discriminator, Relation Guider, to capture the relations between the whole passage and the associated answer and then the Multi-Interaction mechanism is deployed to transfer the knowledge dynamically for our question generation system. Experiments show the effectiveness of our method in both automatic evaluations and human evaluations.

When deploying autonomous agents in the real world, we need effective ways of communicating objectives to them. Traditional skill learning has revolved around reinforcement and imitation learning, each with rigid constraints on the format of information exchanged between the human and the agent. While scalar rewards carry little information, demonstrations require significant effort to provide and may carry more information than is necessary. Furthermore, rewards and demonstrations are often defined and collected before training begins, when the human is most uncertain about what information would help the agent. In contrast, when humans communicate objectives with each other, they make use of a large vocabulary of informative behaviors, including non-verbal communication, and often communicate throughout learning, responding to observed behavior. In this way, humans communicate intent with minimal effort. In this paper, we propose such interactive learning as an alternative to reward or demonstration-driven learning. To accomplish this, we introduce a multi-agent training framework that enables an agent to learn from another agent who knows the current task. Through a series of experiments, we demonstrate the emergence of a variety of interactive learning behaviors, including information-sharing, information-seeking, and question-answering. Most importantly, we find that our approach produces an agent that is capable of learning interactively from a human user, without a set of explicit demonstrations or a reward function, and achieving significantly better performance cooperatively with a human than a human performing the task alone.

Causal approaches to fairness have seen substantial recent interest, both from the machine learning community and from wider parties interested in ethical prediction algorithms. In no small part, this has been due to the fact that causal models allow one to simultaneously leverage data and expert knowledge to remove discriminatory effects from predictions. However, one of the primary assumptions in causal modeling is that you know the causal graph. This introduces a new opportunity for bias, caused by misspecifying the causal model. One common way for misspecification to occur is via unmeasured confounding: the true causal effect between variables is partially described by unobserved quantities. In this work we design tools to assess the sensitivity of fairness measures to this confounding for the popular class of non-linear additive noise models (ANMs). Specifically, we give a procedure for computing the maximum difference between two counterfactually fair predictors, where one has become biased due to confounding. For the case of bivariate confounding our technique can be swiftly computed via a sequence of closed-form updates. For multivariate confounding we give an algorithm that can be efficiently solved via automatic differentiation. We demonstrate our new sensitivity analysis tools in real-world fairness scenarios to assess the bias arising from confounding.

Machine learning plays an increasing role in intelligent tutoring systems as both the amount of data available and specialization among students grow. Nowadays, these systems are frequently deployed on mobile applications. Users on such mobile education platforms are dynamic, frequently being added, accessing the application with varying levels of focus, and changing while using the service. The education material itself, on the other hand, is often static and is an exhaustible resource whose use in tasks such as problem recommendation must be optimized. The ability to update user models with respect to educational material in real-time is thus essential; however, existing approaches require time-consuming re-training of user features whenever new data is added. In this paper, we introduce a neural pedagogical agent for real-time user modeling in the task of predicting user response correctness, a central task for mobile education applications. Our model, inspired by work in natural language processing on sequence modeling and machine translation, updates user features in real-time via bidirectional recurrent neural networks with an attention mechanism over embedded question-response pairs. We experiment on the mobile education application SantaTOEIC, which has 559k users, 66M response data points as well as a set of 10k study problems each expert-annotated with topic tags and gathered since 2016. Our model outperforms existing approaches over several metrics in predicting user response correctness, notably out-performing other methods on new users without large question-response histories. Additionally, our attention mechanism and annotated tag set allow us to create an interpretable education platform, with a smart review system that addresses the aforementioned issue of varied user attention and problem exhaustion.

Active learning agents typically employ a query selection algorithm which solely considers the agent's learning objectives. However, this may be insufficient in more realistic human domains. This work uses imitation learning to enable an agent in a constrained environment to concurrently reason about both its internal learning goals and environmental constraints externally imposed, all within its objective function. Experiments are conducted on a concept learning task to test generalization of the proposed algorithm to different environmental conditions and analyze how time and resource constraints impact efficacy of solving the learning problem. Our findings show the environmentally-aware learning agent is able to statistically outperform all other active learners explored under most of the constrained conditions. A key implication is adaptation for active learning agents to more realistic human environments, where constraints are often externally imposed on the learner.

In this article, I will present several techniques for you to make your first steps towards developing an algorithm that could be used for a classic image classification problem: detecting dog breed from an image. By the end of this article, we'll have developed code that will accept any user-supplied image as input and return an estimate of the dog's breed. Also, if a human is detected, the algorithm will provide an estimate of the dog breed that is most resembling. This project was completed as part of Udacity's Machine Learning Nanodegree (GitHub repo). Convolutional neural networks (also refered to as CNN or ConvNet) are a class of deep neural networks that have seen widespread adoption in a number of computer vision and visual imagery applications.

Higher education institutions share a goal of making learning more accessible to all students. To meet this goal many colleges and universities, including UMass Amherst, have adopted the Universal Design for Learning (UDL) framework in an effort to design curriculum to serve all learners, regardless of ability, disability, age, gender or background. Modern technologies often play a supporting role in UDL, providing students with multiple modalities such as video, audio, and text. While these technologies can make implementing UDL easier, they can also be costly. Beginning with the Fall 2018 term, an interdisciplinary team of academic technologists, instructional designers, and instructors at UMass Amherst started exploring how classroom video and Echo360's new automated speech recognition (ASR) technology can create a pathway to cost-effective, scalable captioning that can improve accessibility and support universal design.

In 2009, her school district introduced an algorithmic method to measure teacher performance, with the plan of terminating the bottom 2% of the district's staff. By 2011, Wysocki had gotten the boot. Because the computer had identified her a bad teacher, despite the raves of parents and administrators. The district's algorithm was based on year-over-year improvements on standardized test scores, and teachers knew this. What became largely apparent was that some teachers, anxious not to be in that bottom 2%, had cheated to inflate scores on their students' behalf.

It's a common sentiment that Deep Learning is only good for images and language models. This post is about using Deep Learning on tabular data, for both Regression and Classification problems. We will use fastai library for creating our deep learning models. We will use Kaggle competitions as benchmarks to see how our solutions compares to other solutions using traditional ML models. If you haven't watched fastai tutorials already, please visit this link for the awesome and free tutorials.