A long-term goal of reinforcement learning is to design agents that can autonomously interact and learn in the world. A critical challenge to such autonomy is the presence of irreversible states which require external assistance to recover from, such as when a robot arm has pushed an object off of a table. While standard agents require constant monitoring to decide when to intervene, we aim to design proactive agents that can request human intervention only when needed. To this end, we propose an algorithm that efficiently learns to detect and avoid states that are irreversible, and proactively asks for help in case the agent does enter them. On a suite of continuous control environments with unknown irreversible states, we find that our algorithm exhibits better sample-and intervention-efficiency compared to existing methods.

Recent research has proposed a series of specialized optimization algorithms for deep multi-task models. It is often claimed that these multi-task optimization (MTO) methods yield solutions that are superior to the ones found by simply optimizing a weighted average of the task losses. In this paper, we perform large-scale experiments on a variety of language and vision tasks to examine the empirical validity of these claims. We show that, despite the added design and computational complexity of these algorithms, MTO methods do not yield any performance improvements beyond what is achievable via traditional optimization approaches. We highlight alternative strategies that consistently yield improvements to the performance profile and point out common training pitfalls that might cause suboptimal results. Finally, we outline challenges in reliably evaluating the performance of MTO algorithms and discuss potential solutions.

High sample complexity has long been a challenge for RL. On the other hand, humans learn to perform tasks not only from interaction or demonstrations, but also by reading unstructured text documents, e.g., instruction manuals. Instruction manuals and wiki pages are among the most abundant data that could inform agents of valuable features and policies or task-specific environmental dynamics and reward structures. Therefore, we hypothesize that the ability to utilize human-written instruction manuals to assist learning policies for specific tasks should lead to a more efficient and better-performing agent. We propose the Read and Reward framework. Read and Reward speeds up RL algorithms on Atari games by reading manuals released by the Atari game developers. Our framework consists of a QA Extraction module that extracts and summarizes relevant information from the manual and a Reasoning module that evaluates object-agent interactions based on information from the manual. An auxiliary reward is then provided to a standard A2C RL agent, when interaction is detected. Experimentally, various RL algorithms obtain significant improvement in performance and training speed when assisted by our design.

Self-supervised representation learning solves auxiliary prediction tasks (known as pretext tasks), that do not require labeled data, to learn semantic representations. These pretext tasks are created solely using the input features, such as predicting a missing image patch, recovering the color channels of an image from context, or predicting missing words, yet predicting this \textit{known} information helps in learning representations effective for downstream prediction tasks. This paper posits a mechanism based on approximate conditional independence to formalize how solving certain pretext tasks can learn representations that provably decrease the sample complexity of downstream supervised tasks. Formally, we quantify how the approximate independence between the components of the pretext task (conditional on the label and latent variables) allows us to learn representations that can solve the downstream task with drastically reduced sample complexity by just training a linear layer on top of the learned representation.

By taking a tiny snippet of real audio - just three seconds is enough - from a person, fraudsters can'clone' the individual's voice using freely available AI tools. By taking a tiny snippet of real audio - just three seconds is enough - from a person, fraudsters can'clone' the individual's voice using freely available AI tools. It's an emergency': your child's voicemail that could be a scam Steps to help combat fraud in which criminals use AI-generated replica of a person's voice to deceive victims T he voicemail from your son is alarming. He has just been in a car accident and is highly stressed. He needs money urgently, although it is not clear why, and he gives you some bank details for a transfer.

On the final day of my visit to Japan, I'm alone and floating in some skyscraper's rooftop hot springs, praying no one joins me. For the last few months, I've been using ChatGPT's Advanced Voice Mode as an AI language tutor, part of a test to judge generative AI's potential as both a learning tool and a travel companion. The excessive talking to both strangers and a chatbot on my phone was illuminating as well as exhausting. I'm ready to shut my yapper for a minute and enjoy the silence. When OpenAI launched ChatGPT late in 2022, it set off a firestorm of generative AI competition and public interest.

Groove AI is a shortcut to a One Man Army or a very smart, experienced personal assistant. The Rise of AI --- has made things so much easier --- we can develop and enhance our own creativity --- everything can be accomplished with AI. From a business or entrepreneurial perspective, or if you are just getting started in a hobby, or in a side hustle, most certainly AI can help! All you need to do is... become good at Prompt Engineering! To learn more about the amazing opportunities AI offers check out AI videos on YouTube, connect with AI personalities on Twitter, read articles on Medium, join groups on LinkedIn and Facebook, or ask me.

In 1982, one of the first 3D games ever released doubled as one of the earliest examples of survival horror. In the pixelated 3D Monster Maze, you not only had to find your way out of a maze but survive being hunted by a T. rex. In the decades since, the dino-horror genre has only grown, from 1999's DinoCrisis to 2016's Far Cry Primal, but dinosaurs have also become more than in-game monsters. We've seen dinosaurs as allies (Yoshi, Pokemon), dinosaurs as attractions (park sims like Zoo Tycoon or Jurassic World), or dinosaurs and their fossils as collectibles (see the in-game markets of Sims or Animal Crossing). The way games have depicted both ancient animals and the paleontologists who study them has gotten richer and deeper as time has passed--though there's still plenty of pixelated T. rexes chomping off people's heads.

We are surrounded by artificial intelligence, or, to put it more accurately, we are surrounded by an atmosphere of technologies. We are incredibly fortunate to have it since it can enrich our lives. The worst pain anyone may experience is losing a child or a close family member. But they can be brought back to us with the aid of Artificial Intelligence. DNA, the fundamental component of cells, is made up of genetic material from both parents, and together with AI, it provides the direction for achieving a purpose.

The career path of Colombian physicist Luisa Fernanda Chaparro Sierra took her from studying the Higgs Boson at CERN, to using similar machine learning techniques to gauge perceptions of crime in the Colombian capital of Bogota. Chaparro, currently a Research Professor at Tecnológico de Monterrey in Monterrey, México, says that after finishing her Phd, she had the opportunity to be part of the DataLab (Laboratorio de Datos) of the Universidad Nacional de Colombia where she used the techniques of handling large databases to help understand the problem of the perception of security in Bogota via machine learning methods. "At CERN, we handled large amounts of data and to differentiate between signal and background; we used supervised machine learning techniques, so I used similar methods and adapted others for the case of perception of security," she says, adding that DataLab was composed of mathematicians, physicists, and engineers with knowledge in programming and statistics. "We used Twitter as our data source and reviewed tweets that talked about security in the city for a year," Chaparro says, "The goal was to design a model that would allow us to quantify something as subjective as perception." The researchers were also hoping to find a relationship between it and real crimes by comparing the results with the databases provided by the National Police.