Mediaboss Marketing

Right now, major tech firms are clamouring to replicate the runaway success of ChatGPT, the generative AI chatbot developed by OpenAI using its GPT-3 large language model. Much like potential game-changers of the past, such as cloud-based Software as a Service (SaaS) platforms or blockchain technology (emphasis on potential), established companies and start-ups alike are going public with LLMs and ChatGPT alternatives in fear of being left behind.

While many of these will succeed some in the mass market and others with niche applications many more will likely fail as the market consolidates.

Who, then, are the companies in good stead to challenge OpenAI?

Googles LaMDA has attracted the most attention from mainstream observers of any LLM outside of GPT-3, but for not quite the same reasons.

Months before ChatGPT exploded into national headlines in late 2022, LaMDA was proving controversial after Google engineer Blake Lemoine was suspended for claiming falsely as became evident that it had developed sentience.

In reality, the LaMDA LLM operates similarly to its main competitor, except that it has fewer parameters at 137 billion compared to 175 billion for GPT-3.5, which was used to train ChatGPT.

LaMDA is also the bedrock of Googles chatbot competitor, named Bard, which the search giant is currently testing for search with select users. Bard had an inauspicious start, however, as it presented a factual error during a launch event.

Israel-based start-up AI21, while less well-known than its rival OpenAI, is a serious challenger in the market. The company created the Jurassic-2 large language model in 2021 with a similar number of parameters to GPT-3.5 178 billion compared to 175 billion and customisation capabilities.

March 2023 then saw the release of Jurassic-2, which focuses on optimised performance as opposed to size. According to AI21, the smallest version of Jurassic-2 outperforms even the largest version of its predecessor. It will also contain a grammatical correction API and text segmentation capabilities.

Users of AI21 studio can train their own versions of the LLM with as few as 50-100 training examples, which then become available for exclusive use.

AI21 also deployed Jurassic-1, and now Jurassic-2 to underpin its WordTune Spices chatbot, which distinguishes itself as a ChatGPT alternative by the use of live data retrieval and the citation of sources in its formulations. Given the risks of factual error and plagiarism associated with LLM chatbots, this is a significant advantage in an increasingly competitive field.

Founded by former OpenAI employees, Anthropic is fast making waves as a rival to its quasi-progenitor.

The generative AI company has launched its own large language model, Claude, whose ChatGPT alternative boasts what it calls constitutional AI. In effect, the model is designed to act according to programmed principles (i.e. its constitution) as opposed to ChatGPT, which is prohibited from answering certain controversial or dangerous queries.

Much like Microsofts investment in OpenAI, Google has invested $300m into Anthropic for a 10% stake in the company.

Baidu Chinas answer to Google is looking to combat its long-term struggles in the face of rival Tencent with its heavy investment in AI.

The team at Baidu has expanded its ERNIE 3.0 large language model into a new version called ERNIE 3.0 Titan. While its predecessor had just 10 billion parameters, Titans PaddlePaddle platform operates on 260 billion.

Titans creators claim that it is the largest dense pre-trained model so far and that it outperforms state-of-the-art models on natural language processing (NLP) tasks.

Hardware and software supplier Nvidia is currently core to the operation of ChatGPT, with an estimated 10,000 of the companys GPUs used to train the chatbot and a predicted 30,000 to be used in future.

This dynamic could be upended, however, as Nvidia CEO Jensen Huang announced in February 2023 that the firm plans to make its DGX AI supercomputer available via the cloud.

Already accessible through Oracle Cloud Infrastructure and Microsoft Azure, the AI supercomputer will have the capacity to allow customers to train their own large language models.

Nvidia has seen a financial boost as companies such as Google and Microsoft look to it for the GPUs necessary for training.

British AI company and Alphabet subsidiary Deepmind, famous for its AlphaGo program, is investing heavily in large language model research and development. Deepmind has iterated on multiple LLMs, including Gopher, Chinchilla and the RETRO system, which combines an LLM with an external database.

This experimentation is leading the way in more targeted and energy-efficient types of LLM Chinchilla has just 70 billion parameters, as opposed to others with double, triple or even more than that, yet outperforms the larger Gopher at certain tasks. Likewise for the 7.5 billion-parameter RETRO, whose external database allows it to outperform vastly larger models.

Not content to invest in the metaverse, Meta has also entered the LLM space with its LLaMA model. Mark Zuckerbergs company does not yet have a publicly available ChatGPT alternative but it is in development.

Unlike many others, the 65-billion parameter LLM has been made open source (upon request, crucially) with the intention of knowledge sharing and crowdsourcing bug fixes.

But just a week after it launched, a torrent for the LLM found its way to the wider internet via a 4Chan leak, prompting fears that such unfettered access could be used for phishing and other cybercrime activities.

Read this article:
These companies are creating ChatGPT alternatives - Tech Monitor

A Google artificial intelligence algorithm on Tuesday inched closer to once again claiming the title of world champion of the ancient Chinese game of Go, besting its human opponent in the first match of a best-of-three championship.

The algorithm, called AlphaGo, is the brainchild of DeepMind, the artificial intelligence research arm of Google parent company Alphabet. It faced off against 19-year old Kie Jie, who is the current human world champion of Go, a strategy game similar to chess that requires players to place black or white stones on a board and capture the opponent's pieces or surround empty spaces to build territories.

"Last year, it was still quite humanlike when it played," Mr. Ke told(Opens in a new window) the New York Times after AlphaGo's win on Tuesday. "But this year, it became like a god of Go."

If the algorithm wins a second game, it will be the second time it has stolen the Go crown from a human opponent. Last year, AlphaGo defeated the previous world champ Lee Sedol in Seoul, Korea. That tournament was a five-game series that saw AlphaGo win the first three matches, although the tournament continued just for fun, with Sedol making a comeback in game four only to be defeated again in the final match.

Go is a strategy game, and its playershuman or otherwisemust frequently adapt and adjust to their opponents' moves. That makes it an ideal challenge for artificial intelligence, which can use machine learning techniques to avoid repeating its own past mistakes, as well as those of its human competitors, as DeepMind CEO Demis Hassabis noted during Tuesday's match.

"Ke Jie is using the ideas AlphaGo used in the master series of online games in January against AlphaGo," Hassabis tweeted(Opens in a new window). "Intriguing to see what it will do."

In the end, the algorithm ended up beating Jie by just half a point, which suggests that the outcome of the final two matches is anyone's guess.

Sign up for What's New Now to get our top stories delivered to your inbox every morning.

This newsletter may contain advertising, deals, or affiliate links. Subscribing to a newsletter indicates your consent to our Terms of Use and Privacy Policy. You may unsubscribe from the newsletters at any time.

Visit link:
Google's AlphaGo AI Beats Human Go Champion | PCMag

The game ofGooriginated in China more than 2,500 years ago.Confuciuswrote about the game, and it is considered one of thefour essential artsrequired of any true Chinese scholar.Played by more than 40 million people worldwide, the rules of the game are simple: Players take turns to place black or white stones on a board, trying to capture the opponent's stones or surround empty space to make points of territory. The game is played primarily through intuition and feel, and because of its beauty, subtlety and intellectual depth it has captured the human imagination for centuries.

But as simple as the rules are, Go is a game of profound complexity. There are 1,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000 possible positionsthats more than the number of atoms in the universe, and more than a googol times larger than chess.

This complexity is what makes Go hard for computers to play, and therefore an irresistible challenge to artificial intelligence (AI) researchers, who use games as a testing ground to invent smart, flexible algorithms that can tackle problems, sometimes in ways similar to humans. The first game mastered by a computer wasnoughts and crosses(also known as tic-tac-toe) in 1952.Then fell checkers in 1994. In 1997Deep Blue famously beat Garry Kasparov atchess. Its not limited to board games eitherIBM'sWatson[PDF] bested two champions at Jeopardy in 2011, andin 2014 our own algorithms learned to play dozens of Atari gamesjust from theraw pixel inputs. But to date, Go has thwarted AI researchers; computers still only play Go as well as amateurs.

Traditional AI methodswhich construct asearch treeover all possible positionsdont have a chance in Go. So when we set out to crack Go, we took a different approach. We built a system, AlphaGo, that combines anadvanced tree searchwithdeep neural networks. These neural networks take a description of the Go board as an input and process it through 12 different network layers containing millions of neuron-like connections. One neural network, the policy network, selects the next move to play. The other neural network, the value network, predicts the winner of the game.

We trained the neural networks on 30 million moves from games played by human experts, until it could predict the human move 57 percent of the time (the previous record before AlphaGo was44 percent). But our goal is to beat the best human players, not just mimic them. To do this, AlphaGo learned to discover new strategies for itself, by playing thousands of games between its neural networks, and adjusting the connections using a trial-and-error process known as reinforcement learning. Of course, all of this requires ahuge amount of computing power, so we made extensive use ofGoogle Cloud Platform.

After all that training it was time to put AlphaGo to the test. First, we held a tournament between AlphaGo and the other top programs at the forefront of computer Go. AlphaGo won all but one of its 500 games against these programs. So the next step was to invite the reigning three-time European Go champion Fan Huian elite professional player who has devoted his life to Go since the age of 12to our London office for a challenge match. In a closed-doors match last October, AlphaGo won by 5 games to 0. It was the first time a computer program has ever beaten a professional Go player. You can find out more in our paper, which was published inNaturetoday.

Originally posted here:
AlphaGo: using machine learning to master the ancient game of Go - Google

Yiqing Xu is a senior studying Computer Science with an interest in a variety of programming languages and a solid math background.

The board game Go has been viewed as one of the most challenging tasks for artificial intelligence because it is complex, pattern-based and hard to program. The computer program AlphaGos victory over Lee Sedol became a huge moment in the history of artificial intelligence and computer engineering. We can observe AlphaGos enormous capacity, but people know little about how it thinks. AlphaGos rules are learned and not designed, implementing machine learning as well as several neural networks to create a learning component and become better at Go. Seen in its partnership with the UKs National Health Service, AlphaGo has promising applications in other realms as well.

From March 9 to March 15 in 2016, a Go game competition took place between the worlds second-highest ranking professional player, Lee Sedol, and AlphaGo, a computer program created by Googles DeepMind Company. AlphaGos 4-1 victory over Lee Sedol became a significant moment in the history of artificial intelligence. This was the first time that a computer had beaten a human professional at Go. Most major South Korean television networks carried the game. In China, 60 million people watched it; the American Go Association and DeepMinds English-language livestream of it on YouTube had 100,000 viewers. A few hundred members of the press watched the game alongside expert commentators [1]. What makes this game so important? To understand this, we have to understand the roots of Go first.

Go, known as weiqi in China and igo in Japan, is an abstract board game for two players that dates back 3,000 years. It is a board game of abstract strategy played across a 19*19 grid. Go starts with an empty board. At each turn, a player places a black or white stone on the board [2]. The general objective of the game is to use the stones to surround more territory than the opponent. Although the rule is very simple, it creates a challenge of depth and nuance. Thus, the board game, Go, has been viewed as one of the most challenging tasks for artificial intelligence because of its complexity and pattern-based state.

In common computer games, the AI usually uses a game tree to determine the best next move in the game depending on what the opponent might do. A game tree is a directed graph that represents game states (positions) as nodes, and possible moves as edges. The root of the tree represents the state at the beginning of the game. The next level represents the possible states after the subsequent moves [3]. Take the simple game tic-tac-toe as an example, it is possible to represent all possible game states visually in Figure 1 [3].

Figure 1: A complete game tree for tic-tac-toe [3].

However, for complex games like Go, getting the best next move in the game quickly becomes impossible since the game tree for Go will contain 10^761 nodes, an overwhelming amount to store in a computer (the universe has only 10^80 atoms, for reference) [4]. This explains why Go has been viewed as one of the greatest challenges for artificial intelligence for so long. Most AIs for board games use hand-crafted rules created by AI engineers. Since these rules might be incomplete, they usually limit the intelligence of the AI. For example, for a certain stage of Go, the designers think the computer should choose one of ten selected steps, but these might be silly moves for professional players. The Go game level of the designers will influence the intelligence level of the AI.

So how did AlphaGo solve the complexity of Go as well as the restriction imposed by the game level of the designers? All previous methods for Go-playing AI relied on some kind of game tree search, combined with hand-crafted rules. AlphaGo, however, makes extensive use of machine learning to avoid using hand-crafted rules and improve efficiency. Machine learning is a type of artificial intelligence that provides computers with the ability to learn without being explicitly programmed. Machine learning focuses on the development of computer programs that can teach themselves to grow and change when exposed to new data.The machine learning systems search through data to look for patterns. But instead of extracting data for human comprehension as is the case in data mining applications it uses the data todetect patterns and adjust program actions accordingly [4]. AlphaGo also uses deep learning and neural networks to teach itself to play. Just like iPhotos is able to help you divide photos into different albums according to different characters because it holds the storage of countless character images that have been processed down to the pixel level, AlphaGos intelligence is based on it having been shown millions of Go positions and moves from human-played games.

AlphaGos intelligence relies on two different components: a game tree search procedure and neural networks that simplify the tree search procedure. The tree search procedure can be regarded as a brute-force approach, whereas the convolutional networks provide a level of intuition to the game-play [5]. The neural networks are conceptually similar to the evaluation function in other AIs, except that AlphaGos are learned and not designed, thus solving the problem of the game level of the designers influencing the intelligence level of AI.

Generally, two main kinds of neural networks inside AlphaGo are trained: policy network and value network. Both types of networks take the current game state as input and grade each possible next move through different formulas and output the probability of a win. On one side, the value network provides an estimate of the value of the current state of the game: what is the probability of the black player to ultimately win the game, given the current state? The output of the value network is the probability of a win. On the other side, the policy networks provide guidance regarding which action to choose, given the current state of the game. The output is a probability value for each possible legal move (the output of the network is as large as the board). Actions (moves) with higher probability values correspond to actions that have a higher chance of leading to a win. One of the most important aspects of AlphaGo is learning-ability. Deep learning allows AlphaGo to continually improve its intelligence by playing a large number of games against itself. This trains the policy network to help AlphaGo predict the next moves, which in turn trains the value network to ascertain and evaluate those positions [5]. AlphaGo looks ahead at possible moves and permutations, going through various eventualities before selecting the one it deems most likely to succeed.

In general, the combined two neural networks let AlphaGo avoid doing excess work: the policy network focuses on the present and decides the next step to save time on searching the entire game tree, and the value network focuses on the long run, analyzing the whole situation to reduce possible moves in the game tree. AlphaGo then averages the suggestion from two networks to make a final decision. What makes AlphaGo so important is that it not only follows the game theory but also involves a learning component. By playing against itself, AlphaGo automatically became better and better at Go.

The Go games were fascinating, but more important is AlphaGos demonstration of the ways artificial intelligence algorithms will affect our lives; AI will make humans better. In the 37th move in the second game, AlphaGo made a very surprising decision. A European Go champion said Its not a human move. Ive never seen a human play this move. So beautiful. This European Go champion who helped teach AlphaGo by playing against it said that though he lost almost all the games, his understanding of Go was greatly improved due to the unusual way the program played. This was also reflected by his jump in world rankings [6].

According to the data, in the United States, there are around 40,500 patients that die of misdiagnosis. The amount of medical information available is huge, so it is impossible for doctors to sort through every little thing. AIs like AlphaGo are able to collect all the medical literature history as well as medical cases, medical images, and other data in the system, and can output the best solution to help doctors. Recently, AlphaGo launched a partnership with the UKs National Health Service to improve the process of delivering care with digital solutions. AlphaGo uses its computing power to analyze health data and records. [6] This will open up new treatment opportunities to patients and assist physicians in treating patients. The increased efficiency will also reduce costs for insurance companies [6].

People already learn so much from the best humans, but now even more knowledge can be acquired from AI. [6] Artificial intelligence can surpass human capabilities in certain situations, and this may make some people uncomfortable. Artificial intelligence uses many techniques in addition to the board game artificial intelligence represented by AlphaGo, with a variety of technical fields including visual recognition and voice recognition. The fact that AI can outperform humans in a specialized area is not surprising. However, in comprehensive intelligence and learning ability, humans are much better than AIs. Although deep learning has made a lot of progress, machine learning still relies on a manual design progress. Moreover, deep learning requires a large amount of data as a basis for training and learning, and the learning process is not flexible enough.

The idea that a comprehensive artificial intelligence will control humans and will have a devastating impact on human society is fictitious. It is not impossible that AI will go beyond human, but that day is still far away, and the beyond will still be under human control.

Whether it is AlphaGo or Lee Sedol winning, overall the victory lies with humankind. The AI behind AlphaGo uses machine learning and neural networks to allow itself to continually improve its skills by playing against itself. This technique of artificial intelligence also offers potential for bettering our lives.

The AI won the Go game, but the human won the future.

[1] How Googles AlphaGo Beat a Go World Champion The Atlantic. Web. 28 Mar. 2016 <http://www.theatlantic.com/technology/archive/2016/03/the-invisible- opponent/475611/>

[2] Go (game) The Wikipedia. Web. <https://en.wikipedia.org/wiki/Go_(game)>

[3] Game Tree The Wikipedia. Web. <https://en.wikipedia.org/wiki/Game_tree>

[4] Definition machine learning The WhatIs. Web. Feb. 2016 <http://whatis.techtarget.com/definition/machine-learning?

[5] Google DeepMinds AlphaGo: How it works The tastehit. Web. 16 Mar. 2016 <https://www.tastehit.com/blog/google-deepmind-alphago-how-it-works/>

[6] AlphaGo Can Shape the Future of Healthcare The TMF. Web. 5 April. 2016 <http://whatis.techtarget.com/definition/machine-learning>

The rest is here:
AI Behind AlphaGo: Machine Learning and Neural Network

In March 2016, Googles AlphaGo program defeated the world champion of board game Go [1]. Go is popular in China and Korea and is 2500 years old [Go game rulesare here]. This was a watershed moment in Artificial Intelligence as experts had not expected a computer to defeat professional Go players for many more decades.

IBMs Deep Blue computer defeated Chess Grandmaster Kasparov in 1996 [2]. However, Go is more complicated than chess. There are 2 metricthat captures a game complexity:

Source: Blogwriter analysis and [2]

To give some perspective, the number of atoms in the universe = 10^80 [2]. Therefore, both chess and Go are almost impossible to beatusing brute force. Go is also 200 times more difficult to solve than chess.

AlphaGo achieved this using a combination of data and algorithms

The initial dataset for AlphaGo consisted of 30Mn board positions from 160,000 real-life games (Dataset A). This was divided into 2 parts training and testing dataset. The training dataset was labelled (i.e. every board position corresponded to an eventual win or loss). AlphaGo then developed models to predict moves of a professional player. These models were tested on the testing dataset and the models correctly predicted the human move 57% of the time (far from perfect but prioralgorithms had achieved a success rate of only 44%) [3].

AlphaGo also keeps playing against itself and generates even more data (Dataset B). It, thus, continues to generate and learn from more data and improves in performance.

A game like Go is essentially deterministic i.e. since all the moves must follow certain rules, a powerful computer can develop a game tree (like the one shown for tic tac toe on the right) for all possible moves and then work backwards to identify the move that has the highest probability of success. Unfortunately, Gos game tree is so large and branches out so much that it is impossible for a computer to do this calculation. [4]

AlphaGo identifies the best move by using 2 algorithms together. The first algorithm (Algorithm X) tries to reduce the breadth of the game while the second algorithm (Algorithm Y) reduces the depth of the game. Algorithm X comes up with possible moves for AlphaGo to play while algorithm Y attaches a value to each of these moves. This eliminates a number of moves that would be impractical (i.e. for which the probability of winning would be almost zero) and, thus, focuses the machines computational poweron moves with higher winning probability

Source:https://commons.wikimedia.org/wiki/File:Tic-tac-toe-full-game-tree-x-rational.jpg

The following schematic explains how AlphaGo uses this combination of data and algorithms to win [Note: this schematic is very bare bones and gives a very high level overview of how AlphaGo works]

Source: Blogwriter

AlphaGos value creation is beyond just its capability to solve a board game. Google acquired DeepMind (creator of AlphaGo) in 2014 for $500Mn [5] and, thus, clearly sees value in a seemingly recreational program. IBMs Watson is a great example of recreational programs becoming mainstream technologies. Watson started off as a computer to play jeopardy but now employs more than 10,000 employees [6]and is beingused in healthcare, digital assistants etc.

The advantage of AlphaGo is that its algorithms are general purpose and not specific to Go [2]. It would be comparatively easy for Google to customize the algorithms to solve other AI challenges as well. The data that AlphaGo generates or has collectedis not useful for other application but the algorithms that power the machines are. Google is already using elements of AlphaGo for incremental improvements in its products like search, image recognition (automatic tagging ofimages inside Google Photos), Google assistant [7].

Google achieved the holy grail in Artificial Intelligence by developing AlphaGo. Its investment in these algorithms (and a seemingly worthless attempt to win at a board game) will pay rich dividends by improving its products like search, Photos, Assistant and self-driving cars as well as by solving other big problems in healthcare, manufacturing etc.

Source:https://xkcd.com/1263/

*******************************************************************************************************************************

Sources:

[1]https://www.theatlantic.com/technology/archive/2016/03/the-invisible-opponent/475611/

[2]https://www.scientificamerican.com/article/how-the-computer-beat-the-go-master/

[3]https://blog.google/topics/machine-learning/alphago-machine-learning-game-go/

[4]https://www.tastehit.com/blog/google-deepmind-alphago-how-it-works/

[5]https://techcrunch.com/2014/01/26/google-deepmind/

[6]https://www.nytimes.com/2016/10/17/technology/ibm-is-counting-on-its-bet-on-watson-and-paying-big-money-for-it.html

[7]http://www.theverge.com/2016/3/14/11219258/google-deepmind-alphago-go-challenge-ai-future

[8]https://bits.blogs.nytimes.com/2014/10/07/ibms-watson-starts-a-parade/

[9]https://www.theregister.co.uk/2017/01/20/its_elementary_ibm_when_is_watson_going_to_make_some_money/

Link:
Google AlphaGo: How a recreational program will change the world