Arthur Smalley

Paul Gardner Allen (January 21, 1953 – October 15, 2018) was an American business magnate, computer programmer, researcher, investor, and philanthropist. He co-founded Microsoft Corporation with childhood friend Bill Gates in 1975, which helped spark the microcomputer revolution of the 1970s and 1980s. Microsoft became the world's largest personal computer software company.[1] Allen was ranked as the 44th-wealthiest person in the world by Forbes in 2018, with an estimated net worth of $20.3 billion at the time of his death.[2][3]

Paul Allen was an entrepreneur, investor, and philanthropist known for co-founding Microsoft with Bill Gates. Allen worked at Microsoft from its founding in 1975 until 1983, shortly after being diagnosed with early-stage Hodgkin’s lymphoma. After stepping away from day-to-day activities at Microsoft, Allen remained on the board of the company until 2000.

Allen left regular work at Microsoft in early 1983 after a Hodgkin lymphoma diagnosis, remaining on its board as vice-chairman. He and his sister, Jody Allen, founded Vulcan Inc. in 1986, a privately held company that managed his business and philanthropic efforts. He had a multi-billion dollar investment portfolio, including technology and media companies, scientific research, real estate holdings, private space flight ventures, and stakes in other sectors. He owned the Seattle Seahawks of the National Football League[5] and the Portland Trail Blazers of the National Basketball Association,[6] and was part-owner of the Seattle Sounders FC of Major League Soccer.[7] In 2000 he resigned from his position on Microsoft's board and assumed the post of senior strategy advisor to the company's management team.

Allen was diagnosed with non-Hodgkin's lymphoma again in 2009 and 2018. After the second diagnosis, he wrote his memoir, Idea Man. Allen passed away on October 15, 2018, at the age of 65, from complications caused by cancer. Allen's philanthropic contributions during his life were more than $2.65 billion, funding a range of projects in conservation, climate science, ocean health, technology, museums, epidemics, neuroscience research, and more.

Paul Allen was born in Seattle on January 21, 1953, to parents Ken and Faye Allen. Ken Allen was associate director of libraries at the University of Washington and Faye Allen was a teacher. Allen had a younger sister called Jody. Both of Allen's parents were avid readers and they encouraged their children to read and develop a range of interests, taking them to museums, libraries, and concerts. Allen's love of science began at an early age and he spent hours reading science fiction and drawing rockets and astronauts. At ten years old, he started a science club for his friends in his basement.

Paul G. Allen’s Art at Christie’s Tops $1.5 Billion, Cracking Records

A museum’s worth of masterworks from the Microsoft co-founder’s collection are offered in a two-part charity sale. Five topped $100 million.

Allen began attending Lakeside School in 1965, developing a passion for computer science and learning to write code. He would spot glitches in computer programs and work on correcting them. While at school, Allen made friends with Bill Gates a fellow computer enthusiast. In 1971 he went to college at Washington State University. He dropped out to found Microsoft with Bill Gates.

Research estimates that the Milky Way could contain over 100 million black holes, including the supermassive black hole at the center of the galaxy, Sagittarius A, which is roughly 4 million times the mass of the sun and approximately 26,000 light-years away from Earth. The first image of a black hole was captured in 2019 by the Event Horizon Telescope (EHT) collaboration. The imaged black hole is at the center of the M87 galaxy, 55 million light-years from Earth. In 2021, astronomers released an updated view of the giant black hole at the center of M87, showing the structure in more detail by imaging in polarized light.

A paper published in Nature on February 19, 2024, reported data on a 17 billion solar mass black hole that grows by a solar mass each day. The quasar, the bright core of a galaxy powered by a black hole known as J0529-4351, is considered the most luminous object detected and the fastest-growing black hole recorded. Originally identified many years ago, the paper presents new observations from the Very Large Telescope (VLT) in Chile demonstrating the size and accretion of the supermassive black hole. J0529-4351 is located 12 billion light-years away from Earth.

Gemma is a family of lightweight, open models from Google built using the same research and technology as the Gemini models. Gemma was developed by Google DeepMind and other teams across the company. The first two Gemma models (Gemma 2B and Gemma 7B) were released on February 21, 2024. Google has stated that Gemma 2B and Gemma 7B offer "best-in-class performance" compared to open models of the same size. Accompanying the model weights, Google also released tools to support developers using Gemma models including a responsible use guide. Users can fine-tune Gemma models on their own data to adapt to specific application needs, such as summarization or retrieval-augmented generation (RAG). Google plans to continue expanding the Gemma family of models, introducing new variants for different applications.

Based on the transformer decoder architecture, Gemma uses similar architectures, data, and training recipes as the Gemini model family. Gemma 2B and Gemma 7B were released with pre-trained and instruction-tuned variants. Google provided a Responsible Generative AI Toolkit, to provide users with guidance and tools for building safe applications using Gemma. The tool kit includes:

• Safety classification: We provide a novel methodology for building robust safety classifiers with minimal examples.
• Debugging: A model debugging tool helps you investigate Gemma's behavior and address potential issues.
• Guidance: You can access best practices for model builders based on Google’s experience in developing and deploying large language models.

Google also provides toolchains for inference and supervised fine-tuning (SFT) across major frameworks including JAX, PyTorch, and TensorFlow through native Keras 3.0. The pre-trained and instruction-tuned Gemma models are designed to run locally on the user's laptop or desktop or through Google Cloud with deployment on Vertex AI and Google Kubernetes Engine (GKE).

Gemma is designed to follow Google's AI principles with automated techniques used to filter out certain personal information from training sets. Google also used fine-tuning and reinforcement learning from human feedback (RLHF) to align the instruction-tuned models with responsible behaviors. Evaluations, including manual red-teaming, automated adversarial testing, and assessments of model capabilities for dangerous activities were conducted prior to the release of Gemma.

Gemma is a family of lightweight, open models from Google built using the same research and technology as the Gemini models. Gemma was developed by Google DeepMind and other teams across the company. The first two Gemma models (Gemma 2B and Gemma 7B) were released on February 21, 2024. Accompanying the model weights, Google also released tools to support developers using Gemma models including a responsible use guide.

Sora is a text-to-video AI model developed by OpenAI that can generate videos up to a minute long based on user prompts. Sora can generate scenes with multiple characters, specific types of motion, details for both the subject and background, and multiple shots within a single generated video with persistent characters and visual style. OpenAI announced Sora on February 15, 2024, making the model available to red teamers to assess potential areas of harm and risks. Upon the announcement, OpenAI also made the model available to visual artists, designers, and filmmakers to gain feedback on performance.

Sora uses a transformer architecture similar to OpenAI's GPT models. At a high level, Sora turns videos into patches by compressing them into a lower-dimensional latent space and decomposing the representation into spacetime patches. The model builds on previous OpenAI research from the Dall-E and GPT models. In particular, using the recaptioning technique from Dall-E 3 that involvedinvolves generative descriptive captions for visual training data.