Generative models of music audio are typically used to generate output based solely on a text prompt or melody. Boomerang sampling, recently proposed for the image domain, allows generating output close to an existing example, using any pretrained diffusion model. In this work, we explore its application in the audio domain as a tool for data augmentation or content manipulation. Specifically, implementing Boomerang sampling for Stable Audio Open, we augment training data for a state-of-the-art beat tracker, and attempt to replace musical instruments in recordings. Our results show that the rhythmic structure of existing examples is mostly preserved, that it improves performance of the beat tracker, but only in scenarios of limited training data, and that it can accomplish text-based instrument replacement on monophonic inputs. We publish our implementation to invite experiments on data augmentation in other tasks and explore further applications.

Planning in complex environments requires an agent to efficiently query a world model to find a feasible sequence of actions from start to goal. Recent work has shown that Large Language Models (LLMs), with their rich prior knowledge and reasoning capabilities, can potentially help with planning by searching over promising states and adapting to feedback from the world. In this paper, we propose and study two fundamentally competing frameworks that leverage LLMs for queryefficient planning. The first uses LLMs as a heuristic within a search-based planner to select promising nodes to expand and propose promising actions. The second uses LLMs as a generative planner to propose an entire sequence of actions from start to goal, query a world model, and adapt based on feedback. We show that while both approaches improve upon comparable baselines, using an LLM as a generative planner results in significantly fewer interactions. Our key finding is that the LLM as a planner can more rapidly adapt its planning strategies based on immediate feedback than LLM as a heuristic. We present evaluations and ablations on Robotouille and PDDL planning benchmarks and discuss connections to existing theory on query-efficient planning algorithms. Planning is the process of determining a sequence of feasible or optimal actions that guide an agent from an initial state to a desired goal state (LaValle, 2006). Planning assumes access to a world model, enabling the agent to simulate and evaluate potential actions without relying on trial-and-error in the real environment. However, in many domains, such as robot task and motion planning, querying the world model is the most computationally expensive step (Kaelbling & Lozano-Pérez, 2013; Garrett et al., 2021). For instance, each query involves running physics or geometric computations or even running a local optimizer. Large language models (LLMs), trained on Internet-scale data, offer multiple opportunities to enable query-efficient planning. Notably, LLMs come with key capabilities such as (1) powerful priors to identify promising states that make progress toward the goal (Ahn et al., 2022), (2) tractable posteriors by easily conditioning on feedback to adaptively choose actions (Lee et al., 2023), and (3) generating complex sequences of actions to plan to the goal (Janner et al., 2021). Recent works leverage one or more such capabilities to design LLM-based agents that solve various decisionmaking tasks (Yao et al., 2022; Shinn et al., 2023b; Huang et al., 2022b; Zhao et al., 2023). However, we show that naively extending such LLM agents to the planning setting becomes quickly intractable. It must not only select among all possible state-action queries but condition on the history of all queries and observations.

Diffusion models can be viewed as mapping points in a high-dimensional latent space onto a low-dimensional learned manifold, typically an image manifold. The intermediate values between the latent space and image manifold can be interpreted as noisy images which are determined by the noise scheduling scheme employed during pre-training. We exploit this interpretation to introduce Boomerang, a local image manifold sampling approach using the dynamics of diffusion models. We call it Boomerang because we first add noise to an input image, moving it closer to the latent space, then bring it back to the image space through diffusion dynamics. We use this method to generate images which are similar, but nonidentical, to the original input images on the image manifold. We are able to set how close the generated image is to the original based on how much noise we add. Additionally, the generated images have a degree of stochasticity, allowing us to locally sample as many times as we want without repetition. We show three applications for which Boomerang can be used. First, we provide a framework for constructing privacy-preserving datasets having controllable degrees of anonymity. Second, we show how to use Boomerang for data augmentation while staying on the image manifold. Third, we introduce a framework for image super-resolution with 8x upsampling. Boomerang does not require any modification to the training of diffusion models and can be used with pretrained models on a single, inexpensive GPU.

There are many facets and levels of drone education. We will focus on the best educational drones for beginner pilots, perhaps best for children. Drones like the UVify OOri and the Ryze Tello have proper education programs centered around them. These platforms teach you some drone hardware basics, then promote critical thinking as you code flight features for the drone. The basics of flight are covered, the machine will hover in place, but you tell it where to go in the sky, just watch out for that wall.

A number of academics and tech entrepreneurs agree: computer intelligence will one day meet and exceed human intelligence. But almost none of them agree on what happens after that. Depending on who you ask, it could be the end of the world or the greatest period of human prosperity we've ever known. I am squarely in the prosperity camp, but Hollywood lends a lot of momentum to those in the doomsday camp. Movies like Terminator, Transcendence, and The Matrix share the archetypal plot point that machines will enslave or kill mankind.

A number of academics and tech entrepreneurs agree: computer intelligence will one day meet and exceed human intelligence. But almost none of them agree on what happens after that. Depending on who you ask, it could be the end of the world or the greatest period of human prosperity we've ever known. I am squarely in the prosperity camp, but Hollywood lends a lot of momentum to those in the doomsday camp. Movies like Terminator, Transcendence, and The Matrix share the archetypal plot point that machines will enslave or kill mankind.

In it, I scaled a mountain, leaping from platform to platform while the environment around me crumbled. I then headed into a tomb, worked through a few puzzles, and triggered a high-octane escape sequence. A year ago, I enjoyed those opening moments immensely. After playing The Legend of Zelda: Breath of the Wild, though, they felt lifeless and stale. From Tomb Raider to Uncharted, the modern adventure game is a tightly choreographed charade, a 20-hour quick time event (QTE) with a clearly defined path. When I jumped to evade an avalanche, Lara landed exactly where the game's developers wanted her to. When I needed to solve a puzzle, the game began pointing, beckoning me to do what the developers wanted me to do.

I'm sure you agree that time is the most precious commodity that none of us can afford to waste. In business, as in life, mastering time management is incredibly important. Clogged and cluttered inboxes drain valuable time and distract you from focusing on what really matters: your work. I've recently tried and tested a number of email tools and techniques in a bid to reclaim valuable time from my working day. As a result, I've genuinely been able to save an hour each day from tedious admin tasks and put that time to much better use.

It's helping medical researchers, aiding in just about every computational process, and beating people in lots of games The AIs Are Winning: 5 Times When Computers Beat Humans The AIs Are Winning: 5 Times When Computers Beat Humans Artificial intelligence is getting good. These aren't AIs that are going to move us closer to the singularity Here's Why Scientists Think You Should be Worried about Artificial Intelligence Here's Why Scientists Think You Should be Worried about Artificial Intelligence Do you think artificial intelligence is dangerous? Boomerang is an add-on for Gmail 5 Smart Addons That Will Make You A Gmail Ninja 5 Smart Addons That Will Make You A Gmail Ninja Gmail has spawned many third party tools, extending it from a mere email service into something much more powerful instead. Siri, Google Now, and Cortana Siri vs Google Now vs Cortana for Home Voice Control Siri vs Google Now vs Cortana for Home Voice Control To find which home voice control is best for you, and which voice assistant fits your specific needs, we've unveiled the pros and cons for Siri, Google Now, and Cortana.

Artificial Intelligence, robots and cognitive computing are hot topics right now. It seems they are part of every conversation as our cars, computers, mobile devices and so many other things in our lives are becoming smarter thanks to new technology. However, there is still a lot of fear, uncertainty and mystery around this new technology. Everything from what it means to peoples' jobs, to whether or not people will be able to create normal relationships in a time where man and machine share the world we live in. Practical applications for AI are in the works, and not all of its uses will be far reaching.