Lifeboat Foundation

The plant-based antiviral agent thapsigargin (TG), derived from a group of poisonous plants known as ‘deadly carrots’, appears to be effective against all variants of SARS-CoV-2 in the lab – and that includes the quick-spreading Delta variant.

A previous study published in February demonstrated that TG can be effective against a host of viruses. Now, this latest work by the same research team confirms that the antiviral also isn’t being outflanked as SARS-CoV-2 evolves. With the emergence of new variants an ongoing possibility, it’s intriguing to observe the continuous efficacy of TG.

In tests on cell cultures in the lab, doses of TG delivered either before infection or during active infection were shown to block and inhibit SARS-CoV-2 variants, triggering a broad and powerful protective response.

Solid-state nuclear magnetic resonance (NMR) spectroscopy—a technique that measures the frequencies emitted by the nuclei of some atoms exposed to radio waves in a strong magnetic field—can be used to determine chemical and 3D structures as well as the dynamics of molecules and materials.

A necessary initial step in the analysis is the so-called chemical shift assignment. This involves assigning each peak in the NMR spectrum to a given atom in the molecule or material under investigation. This can be a particularly complicated task. Assigning chemical shifts experimentally can be challenging and generally requires time-consuming multi-dimensional correlation experiments. Assignment by comparison to statistical analysis of experimental chemical shift databases would be an alternative solution, but there is no such database for molecular solids.

A team of researchers including EPFL professors Lyndon Emsley, head of the Laboratory of Magnetic Resonance, Michele Ceriotti, head of the Laboratory of Computational Science and Modeling and Ph.D. student Manuel Cordova decided to tackle this problem by developing a method of assigning NMR spectra of organic crystals probabilistically, directly from their 2D chemical structures.

NVIDIA recently rolled out a demo of GAUGAN 2, an artificial intelligence-based text to image creation tool. GAUGAN 2 takes keywords and phrases you type in as input, and then generates unique images based on them.

In NVIDIA’s demo video, a user inputs “mountains by a lake” and GAUGAN 2 spits out a beautiful alpine landscape with a small lake in the foreground. We tried using GAUGAN 2 and, in practice, things aren’t as smooth as the demo implies. Certain keywords resulted in bizarre, terrifying results. GAUGAN 2 used this author’s name, for instance, to output an image of what looked like fungi on legs, walking down a street.

GAUGAN 2 is early in development at this point, and likely been trained only on a rather limited data set. Regardless, when it works, it offers a breathtaking snapshot of how AI technology could transform asset creation in movies in games in the years to come, with unique photorealistic landscapes and objects generated from just a few words of user input.

(CNN) — A British man has become the first patient in the world to be fitted with a 3D printed eye, according to Moorfields Eye Hospital in London.

Steve Verze, who is 47 and an engineer from Hackney, east London, was given the left eye on Thursday and first tried it for size earlier this month.

Moorfields Eye Hospital said in a press release Thursday that the prosthetic is the first fully digital prosthetic eye created for a patient.

A new Artificial Intelligence model manages to do complex physics simulations in real time with only using a fraction of the power that a traditionally computed simulation would use. These simulations could soon be used for things like biotechnology, gaming, weather predictions and more. Two Minute Papers has done several videos on it before, but this is a more complex AI with a wider range of applications.
–
TIMESTAMPS:
00:00 The Future of Advanced Physics Simulations.
01:57 How this new approach to AI works.
04:03 Are medical simulations a possibility?
06:02 Last Words.
–
#ai #physics #simulation

A new and revolutionary approach to building Artificial Intelligence models has shown promise of enabling almost any device, regardless of how powerful it is, to run enormous and intelligent Artificial Intelligence’s in a similar way to how our Human Brain operate. This is partially done with new and improved Neuromorphic Computing Hardware which is modelled after our real brains. We may soon see AI beating humans at many different general tasks like an Artificial General Intelligence.
–
TIMESTAMPS:
00:00 The Impossibility of Human AI
01:54 A new Approach is in town.
04:33 Other approaches to AI
06:44 Is this the Future of Artificial Intelligence?
09:43 Last Words.
–
#ai #agi #neuralcomputing

Welcome to AIP.
- The main focus of this channel is to publicize and promote existing SoTA AI research works presented in top conferences, removing barrier for people to access the cutting-edge AI research works.
- All videos are either taken from the public internet or the Creative Common licensed, which can be accessed via the link provided in the description.
- To avoid conflict of interest with the ongoing conferences, all videos are published at least 1 week after the main events. A takedown can be requested if it infringes your right via email.
- If you would like your presentation to be published on AIP, feel free to drop us an email.
- AI conferences covered include: NeurIPS (NIPS), AAAI, ICLR, ICML, ACL, NAACL, EMNLP, IJCAI

If you would like to support the channel, please join the membership:
https://www.youtube.com/c/AIPursuit/join.

Continue reading “Yann LeCun” »

This video introduces the self-supervised representation learning and why we need it. It also discusses the entire pipeline of performing the self-supervised Learning (Learning form unlabeled data). This is the part-1 of the video. The link for part-2 is — https://www.youtube.com/watch?v=j-RSaXzwIHw

Deep Learning systems can achieve remarkable, even super-human performance through supervised learning on large, labeled datasets. However, there are two problems: First, collecting ever more labeled data is expensive in both time and money. Second, these deep neural networks will be high performers on their task, but cannot easily generalize to other, related tasks, or they need large amounts of data to do so. In this blog post, Yann LeCun and Ishan Misra of Facebook AI Research (FAIR) describe the current state of Self-Supervised Learning (SSL) and argue that it is the next step in the development of AI that uses fewer labels and can transfer knowledge faster than current systems. They suggest as a promising direction to build non-contrastive latent-variable predictive models, like VAEs, but ones that also provide high-quality latent representations for downstream tasks.

OUTLINE:
0:00 — Intro & Overview.
1:15 — Supervised Learning, Self-Supervised Learning, and Common Sense.
7:35 — Predicting Hidden Parts from Observed Parts.
17:50 — Self-Supervised Learning for Language vs Vision.
26:50 — Energy-Based Models.
30:15 — Joint-Embedding Models.
35:45 — Contrastive Methods.
43:45 — Latent-Variable Predictive Models and GANs.
55:00 — Summary & Conclusion.

Continue reading “Yann LeCun — Self-Supervised Learning: The Dark Matter of Intelligence (FAIR Blog Post Explained)” »