This article was published as a part of the Data Science Blogathon. A Machine Learning solution to an unambiguously defined business problem is developed by a Data Scientist ot ML Engineer. The Model development process undergoes multiple iterations and finally, a model which has acceptable performance metrics on test data is taken to the production environment. Taking the final chosen model reaching it out to the users is called deployment and there are a few options available to deploy a model. Kubernetes(also called k8s) is one of the open-source tools used for deploying our applications.

A 2019 study conducted by DataChildFutures found that 46% of participating Italian households had AI-powered speakers, while 40% of toys were connected to the internet. More recent research suggests that by 2023 more than 275 million intelligent voice assistants, such as Amazon Echo or Google Home, will be installed in homes worldwide. As younger generations grow up interacting with AI-enabled devices, more consideration should be given to the impact of this technology on children, their rights and wellbeing. AI-powered learning tools and approaches are often regarded as critical drivers of innovation in the education sector. Often recognized for its ability to improve the quality of learning and teaching, AI is being used to monitor students' level of knowledge and learning habits, such as rereading and task prioritization, and ultimately to provide a personalized approach to learning. Knewton is one example of AI-enabled learning software that identifies knowledge gaps and curates education content in line with user needs.

Have you ever looked up to a starry night sky, seen a shooting star and made a wish? Well, look again and look carefully. Are you sure it is a shooting star, or could it be something else? Can you tell for sure? Well, maybe if your wish comes true, then you can tell with certainty that it was a shooting star, no? This Fall semester at New College of Florida, 7 students in the Applied Data Science master's program joined the world-wide effort in analyzing data collected from cameras watching the night skies.

Currently, more and more schools and universities are switching to blended or online education. Due to the spread of the coronavirus, this trend has become even more popular. Meanwhile, it is barely possible to know how students study at home and whether they cheat on online exams. Subsequently, the matter of control is quite critical for schools and universities. AI proctoring systems can solve this issue perfectly.

BrainPOP creates cross-curricular digital content that engages students and supports teachers. We feature animated movies, student creation and reflection tools, learning games, and interactive quizzes to customizable and playful assessments, lesson plans, professional development opportunities, and beyond. With headquarters based in the Flatiron District in NYC, BrainPOP is used in over 40% of US elementary and middle schools, and welcomes millions of monthly site visitors. We are seeking a talented Data Engineer to join our growing team. In this role, you will work closely with senior staff to build new data pipelines and refactoring existing ones that ingest and transform data from a wide array of sources.

As the higher education sector grapples with the "new normal" of the post-pandemic, the structural issues of the recent past not only remain problematic but have been exacerbated by COVID-related disruptions throughout the education pipeline. Navigating the complexity of higher education has always been challenging for students, particularly at underresourced institutions that lack the advising capacity to provide guidance and support. Areas such as transfer and financial aid are notorious black boxes of complexity, where students lacking financial resources and "college knowledge" are too often left on their own to make decisions that may prove costly and damaging down the line. The educational disruptions that many students have faced during the pandemic will likely deepen this complexity by producing greater variations in individual students' levels of preparation and academic histories, even as stressed institutions have less resources to provide advising and other critical student services. Taken together, these challenges will make it all the more difficult to address the equity gaps that the sector must collectively solve. While not a panacea, recent advances in artificial intelligence methodologies such as machine learning can help to alleviate some of the complexity that students and higher education institutions face.

Counterfactual explanations are attracting significant attention due to the flourishing applications of machine learning models in consequential domains. A counterfactual plan consists of multiple possibilities to modify a given instance so that the model's prediction will be altered. As the predictive model can be updated subject to the future arrival of new data, a counterfactual plan may become ineffective or infeasible with respect to the future values of the model parameters. In this work, we study the counterfactual plans under model uncertainty, in which the distribution of the model parameters is partially prescribed using only the first- and second-moment information. First, we propose an uncertainty quantification tool to compute the lower and upper bounds of the probability of validity for any given counterfactual plan. We then provide corrective methods to adjust the counterfactual plan to improve the validity measure. The numerical experiments validate our bounds and demonstrate that our correction increases the robustness of the counterfactual plans in different real-world datasets.

In order for humans to confidently decide where to employ RL agents for real-world tasks, a human developer must validate that the agent will perform well at test-time. Some policy interpretability methods facilitate this by capturing the policy's decision making in a set of agent rollouts. However, even the most informative trajectories of training time behavior may give little insight into the agent's behavior out of distribution. In contrast, our method conveys how the agent performs under distribution shifts by showing the agent's behavior across a wider trajectory distribution. We generate these trajectories by guiding the agent to more diverse unseen states and showing the agent's behavior there. In a user study, we demonstrate that our method enables users to score better than baseline methods on one of two agent validation tasks.

Enabling reinforcement learning (RL) agents to leverage a knowledge base while learning from experience promises to advance RL in knowledge intensive domains. However, it has proven difficult to leverage knowledge that is not manually tailored to the environment. We propose to use the subclass relationships present in open-source knowledge graphs to abstract away from specific objects. We develop a residual policy gradient method that is able to integrate knowledge across different abstraction levels in the class hierarchy. Our method results in improved sample efficiency and generalisation to unseen objects in commonsense games, but we also investigate failure modes, such as excessive noise in the extracted class knowledge or environments with little class structure.

Hierarchical Imitation Learning (HIL) is an effective way for robots to learn sub-skills from long-horizon unsegmented demonstrations. However, the learned hierarchical structure lacks the mechanism to transfer across multi-tasks or to new tasks, which makes them have to learn from scratch when facing a new situation. Transferring and reorganizing modular sub-skills require fast adaptation ability of the whole hierarchical structure. In this work, we propose Dual Meta Imitation Learning (DMIL), a hierarchical meta imitation learning method where the high-level network and sub-skills are iteratively meta-learned with model-agnostic meta-learning. DMIL uses the likelihood of state-action pairs from each sub-skill as the supervision for the high-level network adaptation, and use the adapted high-level network to determine different data set for each sub-skill adaptation. We theoretically prove the convergence of the iterative training process of DMIL and establish the connection between DMIL and Expectation-Maximization algorithm. Empirically, we achieve state-of-the-art few-shot imitation learning performance on the Meta-world \cite{metaworld} benchmark and competitive results on long-horizon tasks of Kitchen environments.