Undergraduate computing classes typically deliver content through passive lectures and require students to write code from scratch. However, students do not always pay attention in lecture and writing code from scratch can be overwhelming for novice students. Students report feeling frustrated when they cannot figure out what is wrong with their code or wait hours to get help from instructional staff. Students from groups that have been historically marginalized are more at risk of failure in introductory courses since they tend to have less prior programming experience. How can we help students succeed in programming courses, especially those without prior programming experience?

Machine Learning (ML) is the branch of Artificial Intelligence in which we use algorithms to learn from data provided to make predictions on unseen data. Recently, the demand for Machine Learning engineers has rapidly grown across healthcare, Finance, e-commerce, etc. According to Glassdoor, the median ML Engineer Salary is $131,290 per annum. In 2021, the global ML market was valued at $15.44 billion. It is expected to grow at a significant compound annual growth rate (CAGR) above 38% until 2029.

I've always been interested in computers because of their ability to help people better understand the world around them. Over the last decade, much of the research done at Google has been in pursuit of a similar vision -- to help people better understand the world around them and get things done. We want to build more capable machines that partner with people to accomplish a huge variety of tasks. Analysis and synthesis tasks, like crafting new documents or emails from a few sentences of guidance, or partnering with people to jointly write software together. We want to solve complex mathematical or scientific problems. Transform modalities, or translate the world's information into any language. Diagnose complex diseases, or understand the physical world. We've demonstrated early versions of some of these capabilities in research artifacts, and we've partnered with many teams across Google to ship some of these capabilities in Google products that touch the lives of billions of users. But the most exciting aspects of this journey still lie ahead! With this post, I am kicking off a series in which researchers across Google will highlight some exciting progress we've made in 2022 and present our vision for 2023 and beyond. I will begin with a discussion of language, computer vision, multi-modal models, and generative machine learning models.

A widely cited explanation for how humans think about trustworthiness posits that people consider three factors, or traits, of a person (or agent) when they evaluate trustworthiness: ability, benevolence, and integrity [20]. It is common practice for intelligent agent researchers to adapt a psychometric inventory of this three-factor model of trustworthiness for assessing users' perceived trustworthiness of agents [19]. In theory, administering the inventory prior to an interaction allows researchers to assess the role of anticipated agent trustworthiness in users' behavior, while post hoc administration allows researchers to assess whether particular elements of an interaction, perhaps an experimental manipulation, impacted users' opinions of the agent. In practice, however, people frequently misuse information when they form judgments and make decisions [11, 17]. For example, a person who is momentarily happy (sad), perhaps from reminiscing about a positive (negative) event from their recent past, is likely to rate their life satisfaction as higher (lower) than if you asked them when they were in a neutral state [25]. Regardless of the saliency of information, the normative approach is to always use it the same way.

Many real-world problems are characterized by a large number of observations, confounding and spurious correlations, partially observable states, and distal, dynamic rewards with hierarchical reward structures. Such conditions make it hard for both animal and machines to learn complex skills. The learning process requires discovering what is important and what can be ignored, how the reward function is structured, and how to reuse knowledge across different tasks that share common properties. For these reasons, the application of standard reinforcement learning (RL) algorithms (Sutton and Barto, 2018) to solve structured problems is often not effective. Limitations of current RL algorithms include the problem of exploration with sparse rewards (Pathak et al., 2017), dealing with partially observable Markov decision problems (POMDP) (Ladosz et al., 2021), coping with large amounts of confounding stimuli (Thrun, 2000; Kim et al., 2019), and reusing skills for efficiently learning multiple task in a lifelong learning setting (Mendez and Eaton, 2020). Standard reinforcement learning algorithms are best suited when the problem can be formulated as a single-task problem in observable Markov decision problem (MDP). Under these assumptions, with complete observability and with static and frequent rewards, deep reinforcement learning (DRL) (Mnih et al., 2015; Li, 2017) has gained popularity due to the ability to learn an approximated Q-value function directly from raw pixel data in the Atari 2600 platform. This and similar algorithms stack multiple frames to derive states of an MDP, and use a basic ɛ-greedy exploration policy. In more complex cases with partial observability and sparse rewards, extensions have been proposed to include more advanced exploration techniques (Ladosz et al., 2022), e.g.

The standard assumption in reinforcement learning (RL) is that agents observe feedback for their actions immediately. However, in practice feedback is often observed in delay. This paper studies online learning in episodic Markov decision process (MDP) with unknown transitions, adversarially changing costs, and unrestricted delayed bandit feedback. More precisely, the feedback for the agent in episode $k$ is revealed only in the end of episode $k + d^k$, where the delay $d^k$ can be changing over episodes and chosen by an oblivious adversary. We present the first algorithms that achieve near-optimal $\sqrt{K + D}$ regret, where $K$ is the number of episodes and $D = \sum_{k=1}^K d^k$ is the total delay, significantly improving upon the best known regret bound of $(K + D)^{2/3}$.

Autonomous navigation and path-planning around non-cooperative space objects is an enabling technology for on-orbit servicing and space debris removal systems. The navigation task includes the determination of target object motion, the identification of target object features suitable for grasping, and the identification of collision hazards and other keep-out zones. Given this knowledge, chaser spacecraft can be guided towards capture locations without damaging the target object or without unduly the operations of a servicing target by covering up solar arrays or communication antennas. One way to autonomously achieve target identification, characterization and feature recognition is by use of artificial intelligence algorithms. This paper discusses how the combination of cameras and machine learning algorithms can achieve the relative navigation task. The performance of two deep learning-based object detection algorithms, Faster Region-based Convolutional Neural Networks (R-CNN) and You Only Look Once (YOLOv5), is tested using experimental data obtained in formation flight simulations in the ORION Lab at Florida Institute of Technology. The simulation scenarios vary the yaw motion of the target object, the chaser approach trajectory, and the lighting conditions in order to test the algorithms in a wide range of realistic and performance limiting situations. The data analyzed include the mean average precision metrics in order to compare the performance of the object detectors. The paper discusses the path to implementing the feature recognition algorithms and towards integrating them into the spacecraft Guidance Navigation and Control system.

Distributed online optimization and game have been increasingly researched in the last decade, mostly motivated by its wide applications in sensor networks, robotics (e.g., distributed target tracking and formation control), smart grids, deep learning, and so forth. In these problems, there is a network of agents who may be cooperative (i.e., distributed online optimization) or noncooperative (i.e., online game) through local information exchanges. And the local cost function of each agent is often time-varying in dynamic and even adversarial environments. At each time, a decision must be made by each agent based on historical information at hand without knowing future information on cost functions. For these problems, a comprehensive survey is still lacking. This paper aims to provide a thorough overview of distributed online optimization and game from the perspective of problem settings, communication, computation, algorithms, and performances. In addition, some potential future directions are also discussed.

Companies struggle to continuously develop and deploy AI models to complex production systems due to AI characteristics while assuring quality. To ease the development process, continuous pipelines for AI have become an active research area where consolidated and in-depth analysis regarding the terminology, triggers, tasks, and challenges is required. This paper includes a Multivocal Literature Review where we consolidated 151 relevant formal and informal sources. In addition, nine-semi structured interviews with participants from academia and industry verified and extended the obtained information. Based on these sources, this paper provides and compares terminologies for DevOps and CI/CD for AI, MLOps, (end-to-end) lifecycle management, and CD4ML. Furthermore, the paper provides an aggregated list of potential triggers for reiterating the pipeline, such as alert systems or schedules. In addition, this work uses a taxonomy creation strategy to present a consolidated pipeline comprising tasks regarding the continuous development of AI. This pipeline consists of four stages: Data Handling, Model Learning, Software Development and System Operations. Moreover, we map challenges regarding pipeline implementation, adaption, and usage for the continuous development of AI to these four stages.

In natural language processing (NLP), code-mixing (CM) is a challenging task, especially when the mixed languages include dialects. In Southeast Asian countries such as Singapore, Indonesia, and Malaysia, Hokkien-Mandarin is the most widespread code-mixed language pair among Chinese immigrants, and it is also common in Taiwan. However, dialects such as Hokkien often have a scarcity of resources and the lack of an official writing system, limiting the development of dialect CM research. In this paper, we propose a method to construct a Hokkien-Mandarin CM dataset to mitigate the limitation, overcome the morphological issue under the Sino-Tibetan language family, and offer an efficient Hokkien word segmentation method through a linguistics-based toolkit. Furthermore, we use our proposed dataset and employ transfer learning to train the XLM (cross-lingual language model) for translation tasks. To fit the code-mixing scenario, we adapt XLM slightly. We found that by using linguistic knowledge, rules, and language tags, the model produces good results on CM data translation while maintaining monolingual translation quality.