Richard Larson

An interesting study today on the value of open source software: https://lnkd.in/gUfkGmV3 From the abstract: "We estimate the supply-side value of widely-used OSS is $4.15 billion, but that the demand-side value is much larger at $8.8 trillion. We find that firms would need to spend 3.5 times more on software than they currently do if OSS did not exist. The top six programming languages in our sample comprise 84% of the demand-side value of OSS. Further, 96% of the demand-side value is created by only 5% of OSS developers." As a (heavy) user and contributor, it's good to see its value estimated, as free stuff usually goes under appreciated by decision makers.

26

1 Comment

I was recently teaching Physics to fresh class 11th students and wondering if there is another way to prove the volume of a parallelepiped, without using the differential/thin plates analogy… Volume of a Parallelepiped = (AxB).C = Scalar Triple Product https://lnkd.in/ggfdfzZS

New GAI Textbook Announcement (The textbook is being used by Stanford CS372: AI for precision medicine and psychological disorders) The past two years have been a significant period in the field of AI, during which we have observed substantial strides in generative AI technologies. Having dedicated over 5,000 hours to research and analysis, I have compiled this book to share a collection of findings and hypotheses regarding Large Language Models (LLMs)—detailing their operational mechanisms, and strategies to mitigate biases and reduce instances of hallucination. The central premise is to enhance context awareness and define human objectives with greater precision. By doing so, LLMs can more effectively mirror human linguistic behaviors, utilizing context-specific linguistic features to fulfill intended goals. SocraSynth: Socratic Synthesis with Multiple Large Language Models. https://lnkd.in/giC4474p

25

CMU Researchers Propose a Distributed Data Scoping Method: Revealing the Incompatibility between the Deep Learning Architecture and the Generic Transport PDEs Researchers from Carnegie Mellon University present a data scoping technique to augment the generalizability of data-driven models forecasting time-dependent physics properties in generic transport issues by disentangling the expressiveness and local dependency of the neural operator. They solve this problem by suggesting a distributed data scoping approach with linear time complexity, strictly constraining information scope to predict local properties. Numerical experiments across various physics domains demonstrate that their data scoping technique significantly hastens training convergence and enhances the benchmark models’ generalizability in extensive engineering simulations. They outline a generic transport system’s domain in d-dimensional space. Introducing a nonlinear operator evolving the system, aiming to approximate it via a neural operator trained using observations from a probability measure. The discretization of functions allows for mesh-independent neural operators in practical computations. The physical information in a generic transport system travels at a limited speed, and they defined the local-dependent operator for the generic transport system. They also clarify how the deep learning structure of neural operators dilutes local dependency. A neural operator comprises layers of linear operators followed by non-linear activations. As layers increase to capture nonlinearity, the local-dependency region expands, potentially conflicting with time-dependent PDEs’ local nature. Instead of limiting the scope of the linear operator to one layer, they directly limit the scope of input data. The data scoping method decomposes the data so that each operator only works on the segmentation. Quick read: https://lnkd.in/g3ERrWzq Paper: https://lnkd.in/g9GeqBDA Machine Learning Department at CMU #ai

19

Joint work with Aby Philip, Soorya Rethinasamy, and Vincent Russo now published in Quantum - the open journal for quantum science: https://lnkd.in/giwDBREV Title: Schrödinger as a Quantum Programmer: Estimating Entanglement via Steering Popular Summary: Entanglement is a unique feature of quantum mechanics, initially brought to light by Einstein, Podolsky, and Rosen. Determining whether a general quantum state is entangled or not, known as the separability problem, is of fundamental interest and relevant to various fields of physics: condensed matter, quantum gravity, quantum optics, and quantum information science. Motivated by the rapid development of quantum computers, we provide a variational quantum steering algorithm (VQSA) to address the separability problem. The quantum steering effect, discovered by Schrödinger, is central to the design of our algorithm, which is implementable on quantum computers that are available today. We provide benchmarks and simulations of our algorithm and establish a meaningful connection between steering, entanglement, quantum algorithms, and quantum computational complexity theory.

77

4 Comments

I am proud to share that our paper, "Greedy Recursive Spectral Bisection for Modularity-Bound Hierarchical Divisive Community Detection," has been published in Statistics and Computing. Thanks to an agreement between Springer and FCT - Portugal, you can access the full article for free! Dive into our innovative approach for community detection and explore the detailed methodology and results. Check it out here: https://rdcu.be/dL8bM #StatisticsAndComputing #CommunityDetection #Research #OpenAccess #Springer #PPCIC #PPPRO #CEFETRJ #DataScience

30

If you wonder how RNNs can harness noise to perform useful computations 🧠 and how one could analyze a phase transition in random RNNs through a topological field theory approach, check out my recent talk here 👇 https://lnkd.in/gbSPC8Rq

11

AI is transforming the world; Transformers are revolutionizing AI. If you're working in this exciting field, consider submitting your best papers to the ACM Transactions on Intelligent Systems and Technology Special Issue on Transformers.  EiC: Huan Liu; Guest Editors: Feng Xia, Tyler Derr, Anh Tuan Luu, Richa Singh, Aline Villavicencio. #AI #Transformer #LLM #FoundationModel #MachineLearning #DeepLearning #NeuralNetworks #CFP #callforpapers #SpecialIssue #IntelligentSystems #artificialintelligence

83

In our new paper with Airat Kamaletdinov entitled "Coupling Light with Matter for Identifying Dominant Subnetworks" (arXiv:2405.17296), we present a novel light-matter platform that uses complex-valued oscillator networks, a form of physical neural networks, to identify dominant subnetworks and uncover indirect correlations within larger networks. This approach offers potential advantages, including low energy consumption, high processing speed, and the immediate identification of co- and counter-regulated nodes without post-processing. The effectiveness of this approach is demonstrated through its application to biological networks, and we also propose its applicability to a wide range of other network types. https://lnkd.in/ePmYmNX4 Research is funded by HeIsinberg EU Horizon under GA No: 101114978: https://lnkd.in/e_i_9eba

34

Google Unveils Willow: The Future of Quantum Computing Google’s new quantum device, Willow, is a game-changer. To put it simply: What takes a classical computer 10,000 years, Willow does in seconds. Here's how it could impact different areas in simple terms: Faster Medicines: Design new drugs in days, speeding up cures for diseases. Better Weather Predictions: Accurately forecast storms and combat climate change. Safer Internet: Build stronger security to keep your data safe. Smarter AI: Create AI that learns and evolves much faster, improving everything from personal assistants to self-driving cars. Traffic-Free Cities: Optimize city traffic for smoother travel and fewer jams. Cleaner Energy: Design better energy solutions like advanced solar panels and nuclear fusion. Cheaper Flights and Deliveries: Optimize routes to save time, money, and resources. And here's the big one: Understanding the Universe: Simulate the origins of the universe, like the Big Bang, or explore the mysteries of black holes, helping us understand how everything began. Willow isn’t just a beginning of faster chip solving problems—it’s helping us unlock the secrets of life and the universe itself! quantum computing is here.. Dr. Rajeev Tiwari Dr.Taruna Gautam Nihar Amoncar PhD PUNEET SHARMA #QuantumComputing #GoogleWillow #Innovation #FutureTech https://lnkd.in/gY3UFWpm

23

2 Comments

Attish Bhatia’s recent NYT article, "When AI's Output is a Threat to AI Itself," explores the risks of AI systems trained on AI-generated content, potentially leading to "AI collapse." As the web floods with AI-produced text and images, the need for fresh, unbiased data is critical. Read more https://lnkd.in/gtux_F89 #AI #MachineLearning #DataEthics #ArtificialIntelligence #TechChallenges #AIImpact

1

With this Sadler's Lectures podcast episode, we finish up the "Solution of the General Problem of the Prolegomena" in Kant's work, looking at why he thinks appeals to probability or sound common sense won't get you a genuine metaphysics https://lnkd.in/gFvctMkQ #Kant #Podcast #Metaphysics #Philosophy #CommonSense #Probability #Epistemology

1

Here, we propose a weighted normalized gradient-based algorithm that optimizes the sum-of-squared difference between the weighted normalized gradients of the reference and deformed images. Read more here: https://lnkd.in/e6A6HXwM   #Digitalimagecorrelation #Inversecompositionalimagealignment #Normalizedgradient #Imagematching #Illuminationvariation Mechanics of Composites for Energy and Mobility - KAUST KAUST PSE KAUST Research

22

Predictions based on Machine Learning and Statistics often rely on maximal likelihood, which can trap LLMs in biases inherited from the majority opinions of training data. This problem has been substantially addressed through conditional statistics, allowing for mitigation of biases and preventing the cancellation of minority perspectives. https://lnkd.in/gMXKqwyB

7

A new video that introduces a pdf of a brief, Geometric Algebra solution to a contest problem: https://lnkd.in/gbN-VZax

4

I've made a free conversion of ​​"The Para-Academic Handbook A Toolkit for Making-Learning-Creating-Acting" in ePub 3.0 format. Available on github: https://lnkd.in/eTfNCFau The book is a collection of writing from people engaged in para-academic work across a number of disciplines, and gives ideas and inspiration for navigating life on the margins of the academy. This work provides a glimpse into a number of responses to UK austerity measures in the 2010s, but will be of value for today's scholars, whether past, present, or even future members of academe. This conversion is offered under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International,  in keeping with the licensing of the original work. Please grab your copy, and spread the word to others who might benefit!

8

1 Comment

I agree with Jensen that the next wave of AI is physical understanding, especially the ability to model and simulate physics is essential. We’re teaching A.I. physics. Due to the massive amount of data available on the internet, the first wave of generative AI has been digital, but the next wave will be physical. Language models lack physical grounding. They only have a high level of understanding but cannot simulate physical phenomena. Image and video models focus on "looking good" vs. being physically valid. This means that the notion of digital twins needs an upgrade. They need to be trained on laws of physics to have the right foundation. Having physically accurate digital twins means faster design cycles, leading to a huge reduction in R&D costs. Watch my TED talk https://lnkd.in/gCN3WFkP #CES #Nvidia #NVIDIAcosmos

905

71 Comments

We're getting to the end of Kant's Prolegomena To Any Future Metaphysics in the Sadler's Lectures podcast. Here's an episode that takes us into the "Solution Of The General Problem" https://lnkd.in/gCrc2qFr #Podcast #Kant #Metaphysics #Prolegomena #Philosophy #Critique #Reason

3