Reducing domain divergence is a key step in transfer learning problems. Existing works focus on the minimization of global domain divergence. However, two domains may consist of several shared subdomains, and differ from each other in each subdomain. In this paper, we take the local divergence of subdomains into account in transfer. Specifically, we propose to use low-dimensional manifold to represent subdomain, and align the local data distribution discrepancy in each manifold across domains. A Manifold Maximum Mean Discrepancy (M3D) is developed to measure the local distribution discrepancy in each manifold. We then propose a general framework, called Transfer with Manifolds Discrepancy Alignment (TMDA), to couple the discovery of data manifolds with the minimization of M3D. We instantiate TMDA in the subspace learning case considering both the linear and nonlinear mappings. We also instantiate TMDA in the deep learning framework. Extensive experimental studies demonstrate that TMDA is a promising method for various transfer learning tasks.

An important goal in reinforcement learning is to create agents that can quickly adapt to new goals while avoiding situations that might cause damage to themselves or their environments. One way agents learn is through exploration mechanisms, which are needed to discover new policies. However, in deep reinforcement learning, exploration is normally done by injecting noise in the action space. While performing well in many domains, this setup has the inherent risk that the noisy actions performed by the agent lead to unsafe states in the environment. Here we introduce a novel approach called Meta-Learned Instinctual Networks (MLIN) that allows agents to safely learn during their lifetime while avoiding potentially hazardous states. At the core of the approach is a plastic network trained through reinforcement learning and an evolved "instinctual" network, which does not change during the agent's lifetime but can modulate the noisy output of the plastic network. We test our idea on a simple 2D navigation task with no-go zones, in which the agent has to learn to approach new targets during deployment. MLIN outperforms standard meta-trained networks and allows agents to learn to navigate to new targets without colliding with any of the no-go zones. These results suggest that meta-learning augmented with an instinctual network is a promising new approach for safe AI, which may enable progress in this area on a variety of different domains.

To build Sounding Board, we develop a system architecture that is capable of accommodating dialog strategies that we designed for socialbot conversations. The architecture consists of a multi-dimensional language understanding module for analyzing user utterances, a hierarchical dialog management framework for dialog context tracking and complex dialog control, and a language generation process that realizes the response plan and makes adjustments for speech synthesis. Additionally, we construct a new knowledge base to power the socialbot by collecting social chat content from a variety of sources. An important contribution of the system is the synergy between the knowledge base and the dialog management, i.e., the use of a graph structure to organize the knowledge base that makes dialog control very efficient in bringing related content to the discussion. Using the data collected from Sounding Board during the competition, we carry out in-depth analyses of socialbot conversations and user ratings which provide valuable insights in evaluation methods for socialbots. We additionally investigate a new approach for system evaluation and diagnosis that allows scoring individual dialog segments in the conversation. Finally, observing that socialbots suffer from the issue of shallow conversations about topics associated with unstructured data, we study the problem of enabling extended socialbot conversations grounded on a document. To bring together machine reading and dialog control techniques, a graph-based document representation is proposed, together with methods for automatically constructing the graph. Using the graph-based representation, dialog control can be carried out by retrieving nodes or moving along edges in the graph. To illustrate the usage, a mixed-initiative dialog strategy is designed for socialbot conversations on news articles.

Vision-and-Language Navigation (VLN) requires an agent to follow natural-language instructions, explore the given environments, and reach the desired target locations. These step-by-step navigational instructions are crucial when the agent is navigating new environments about which it has no prior knowledge. Most recent works that study VLN observe a significant performance drop when tested on unseen environments (i.e., environments not used in training), indicating that the neural agent models are highly biased towards training environments. Although this issue is considered as one of the major challenges in VLN research, it is still under-studied and needs a clearer explanation. In this work, we design novel diagnosis experiments via environment re-splitting and feature replacement, looking into possible reasons for this environment bias. We observe that neither the language nor the underlying navigational graph, but the low-level visual appearance conveyed by ResNet features directly affects the agent model and contributes to this environment bias in results. According to this observation, we explore several kinds of semantic representations that contain less low-level visual information, hence the agent learned with these features could be better generalized to unseen testing environments. Without modifying the baseline agent model and its training method, our explored semantic features significantly decrease the performance gaps between seen and unseen on multiple datasets (i.e. R2R, R4R, and CVDN) and achieve competitive unseen results to previous state-of-the-art models. Our code and features are available at: https://github.com/zhangybzbo/EnvBiasVLN

Since AI is a branch of Computer Science, I think we face the same issues regarding math as other areas of CS. On the one hand, mathematical concepts are foundational to CS. On the other hand, we want to make the material accessible to as broad a range of students as possible, including those with weak math backgrounds and those who are disinclined toward mathematical thinking but might still find other aspects of computer science attractive. Probability and statistics are two topics that are seen as increasingly important for K-12 students. In AI, they form the basis of probabilistic reasoning (part of our Big Idea #2).

Microsoft AI for Accessibility is funding the Object Recognition for Blind Image Training (ORBIT) project, led by City, University of London's Dr Simone Stumpf. Currently, the project is recruiting blind and low vision users in the UK to record videos of things that are important to them. The collected video data will enable the team to construct a large data set from users who are blind or have low vision which can be used for training and testing AI models that personalise object recognition - and ultimately help build better AI for everyone. Collecting this video data from users who are blind and low vision is, as Dr Stumpf notes, a "tricky process", because it, "must be simultaneously easy for blind users to record the videos and the data must be useful for machine learning." Experience from a pilot study showed users were able to take videos in different settings in their home using different filming techniques.

Of late, few things make me cringe more than the mischaracterization of the word "data," particularly with regards to its variants in Data Science, Data Analytics, Big Data, and the like. However, I don't fault the perpetrators: data's influence on your life is increasing at a rate much faster than that at which efforts are being made to educate the majority of people on what it is and how it is translated into actionable insights. At this point, most of us realize that properly-leveraged data has many lucrative use-cases. In turn, businesses and organizations are seeking to tap into this movement without always knowing what its tangible benefits will be for their specific operations. I don't mean to echo any Revere-esque rhetoric to the tune of "The Data are coming!

This book provides an introduction to statistical learning methods. It is aimed for upper level undergraduate students, masters students and Ph.D. students in the non-mathematical sciences. The book also contains a number of R labs with detailed explanations on how to implement the various methods in real life settings, and should be a valuable resource for a practicing data scientist.

If you are thinking to learn a new programming language then also Python is a good choice, particularly if you are looking to move towards a lucrative career path of Data Science and Machine learning which has lots of opportunities. In this article, I am going to share some of the best online courses to learn Python in 2020... Python is an object-oriented, high-level programming language with integrated dynamic semantics primarily for web and app development. It is extremely attractive in the field of Rapid Application Development because it offers dynamic typing and dynamic binding options. Python is relatively simple, so it's easy to learn since it requires a unique syntax that focuses on readability. Developers can read and translate Python code much easier than other languages.

New Delhi, Jan 11 (IANS) When it comes to programming languages in India, Python is most popular among the students for its role in Artificial Intelligence (AI) applications, data science, Machine Learning (ML) and data analytics, US-based online education company Coursera has said. Python dominated the top 10 list with courses like "Programming for Everybody", "Python Data Structures", "Python for Data Science and AI" and more. Python is also easy to get started with, offers a lot of flexibility and is versatile. "Its open source nature makes it easy to learn. A large number libraries for tasks like web development, text processing, calculations add to its appeal," the repor said.