Connect with us

Big Data

Machine learning systems are a ‘land rush’ of opportunity for CIOs

Published

on

CAMBRIDGE, Mass. — Machine learning is overtaking big data in Google searches, but the hype around artificial intelligence systems may not be hyped enough.

Erik Brynjolfsson and Andrew McAfee, co-authors of the forthcoming Machine, Platform, Crowd: Harnessing Our Digital Future, said some machine learning algorithms are improving faster than anticipated, thanks to enormous data sets and access to more compute power. And the rapid rate of change is opening up new avenues for machines and new opportunities for CIOs.

Brynjolfsson, director of the MIT Initiative on the Digital Economy (IDE), and McAfee, principal scientist and co-director of the MIT IDE, participated in a fireside chat at the recent MIT Sloan CIO Symposium, where they said they underestimated the disruptive capabilities of automation and discussed what they called the second wave of the second machine age. Instead of humans codifying knowledge for machines, the machines are capable of learning on their own.

“We think it’s probably the most important thing that’s affecting the economy and society over the coming decade,” Brynjolfsson said.

A runaway example, McAfee said, is AlphaGo, a system built by Google DeepMind that bested the world champion of the Asian abstract strategy game Go. “The Chinese Go champion tweeted out on the Chinese equivalent of Twitter a little while back, ‘I don’t think that a single human has touched the edge of the game of Go,'” he said. In less than five years, machines may have exceeded 2,500 years of study and accumulated knowledge of the game by humans.

And improvement in machine learning systems isn’t limited to game strategy. Speech and image recognition technologies, for example, are operating with human-level precision and learning at an unprecedented rate. Google revealed at the Google I/O 2017 developer conference that its speech recognition technology improved from an 8% error rate to about a 4% error rate — not in 10 years, but in 10 months, Brynjolfsson said.

Those advances are positioning machine learning systems as the powerhouse of the software world, sitting below the user interface of every chat bot, autonomous vehicle and facial recognition system. And the pace at which machine learning systems are evolving presents a unique situation for CIOs, who shouldn’t sit passively by and let the technology lead, according to the researchers.

Indeed, Brynjolfsson said companies like Microsoft and Google and renowned machine learning experts like Andrew Ng describe what’s happening as “almost a land rush.”

“There are so many opportunities that people haven’t cashed in on yet,” Brynjolfsson said. “And the bottleneck now is actually identifying the problems and opportunities that these technologies can be applied to most effectively.”

Experimentation — such as Google turning a gaming algorithm loose on its data center to reduce power consumption — is key. “The main error that a lot of companies are going to make is to extrapolate from the past and keep doing what they were doing with a little bit better accuracy or a little bit better precision,” McAfee said.

 MIT Sloan CIO Symposium, machine learning, machine learning systems, Jason Pontin, Erik Brynjolfsson, Andrew McAfee
From left to right: Jason Pontin, Erik Brynjolfsson and Andrew McAfee

Machine learning systems and the workforce

But grappling with machine learning systems is not just about learning how to exploit the technology. It will require rethinking whole swaths of employment, from menial to highly skilled jobs.

In The Second Machine Age, published in 2014, the researchers wrote about the “the great decoupling” of wages and productivity. For 200 years, [median wages] rose right alongside productivity, but now they are stagnating,” Brynjolfsson said. With dire effects. Brynjolfsson pointed to Anne Case and Angus Deaton’s research on the rise of deaths from suicide, drug abuse, alcoholism and depression — or, as he put it, “deaths from despair.”

And with advances in machine learning systems, it is not just rote tasks that are prime targets for automation. Jobs that require years of training are also at risk, such as pattern recognition or image analysis done by pathologists and radiologists. “Machines can now recognize potentially cancerous images as well or better than humans can,” Brynjolfsson said.

Just how much of an impact machine learning systems will have on jobs and careers is hard to predict. When Jason Pontin, editor in chief and publisher of the MIT Technology Review and the moderator of the fireside chat, asked the academics to put a percentage on the number of jobs that could be eliminated in the next 10 to 15 years, Brynjolfsson responded by saying, “That’s the multitrillion-dollar question.”

One thing is certain, the authors added: Machines aren’t good at everything. Tasks that require creative thinking, interpersonal connections, large-scale problem solving or complex planning are still better performed by humans — at least for now. And advances in machine learning systems do offer a career opportunity, especially for people with computer smarts.

In the IT space, forward-looking CIOs will consider how to arm their employees with the skills necessary to remain relevant in the machine learning economy, such as learning TensorFlow, the open source software library for machine learning developed by Google, Brynjolfsson said. 

PlatoAi. Web3 Reimagined. Data Intelligence Amplified.

Click here to access.

Source: https://searchcio.techtarget.com/opinion/Machine-learning-systems-are-a-land-rush-of-opportunity-for-CIOs

Big Data

WHT: A Simpler Version of the fast Fourier Transform (FFT) you should know

Published

on

WHT: A Simpler Version of the fast Fourier Transform (FFT) you should know

The fast Walsh Hadamard transform is a simple and useful algorithm for machine learning that was popular in the 1960s and early 1970s. This useful approach should be more widely appreciated and applied for its efficiency.


By Sean O’Connor, a science and technology author and investigator.

The fast Walsh Hadamard transform (WHT) is a simplified version of the Fast Fourier Transform (FFT.)

The 2-point WHT of the sequence a, b is just the sum and difference of the 2 values:

WHT(a, b) = a+b, a-b. 

It is self-inverse allowing for a fixed constant:

WHT(a+b, a-b) = 2a, 2b 

Due to (a+b) + (a-b) = 2a and (a+b) – (a-b) = 2b.

The constant can be split between the two Walsh Hadamard transforms using a scaling factor of √2 to give a normalized WHTN:

WHTN(a, b) = (a+b)/√2, (a-b)/√2 WHTN((a+b)/√2, (a-b)/√2) = a, b 

That particular constant results in the vector length of a, b being unchanged after transformation since a2+b2 =((a+b)/√2)2+ ((a-b)/√2)2 as you may easily calculate.

The 2-point transform can be extended to longer sequences by sequentially adding and subtracting pairs of similar terms, alike in the pattern of + and – symbols they contain.

To transform a 4-point sequence a, b, c, d first do two 2-point transforms:

WHT(a, b) = a+b, a-b WHT(c, d) = c+d, c-d 

Then add and subtract the alike terms a+b and c+d:

WHT(a+b, c+d) = a+b+c+d, a+b-c-d 

and the alike terms a-b and c-d:

WHT(a-b, c-d) = a-b+c-d, a-b-c+d 

The 4-point transform of a, b, c, d then is

WHT(a, b, c, d) = a+b+c+d,  a+b-c-d, a-b+c-d, a-b-c+d 

When there are no more similar terms to add and subtract, that signals completion (after log2(n) stages, where n is 4 in this case.)  The computational cost of the algorithm is nlog2(n) add/subtract operations, where n, the size of the transform, is restricted to being a positive integer power of 2 in the general case.

If the transform was done using matrix operations, the cost would be much higher (n2 fused multiply-add operations.)

Figure 1.  The 4-point Walsh Hadamard transform calculated in matrix form.

The +1, -1 entries in Figure 1 are presented in a certain natural order which most of the actual algorithms for calculating the WHT result in, which is fortunate since then the matrix is symmetric, orthogonal and self-inverse.

You can also view the +1, -1 patterns of the WHT as waveforms.

Figure 2.  The waveforms of the 8-point WHT presented in natural order.

When you calculate the WHT of a sequence of numbers, you are really just determining how much of each waveform is embedded in the original sequence.  And that is complete and total information with which you can fully reconstruct any sequence from its transform.

The waveforms of the WHT typically correlate strongly with the patterns found in natural data like images, allowing the transform to be used for data compression.

Figure 3.  A 65536-pixel image compressed to 5000 points using a WHT.

In Figure 3, a 65536-pixel image was transformed with a WHT, the 5000 maximum magnitude embeddings were preserved, and then the inverse transform was applied (simply another WHT.)

The central limit theorem (CLT) tells you that adding a large quantity of random numbers results in the Normal distribution with its characteristic bell curve.  The CLT applies equally to sums and differences of a large quantity of random numbers.  As a result, C.M. Rader proposed (in 1969) using the WHT to quickly generate Normally distributed random numbers from conventional uniformly distributed random numbers.  You simply generate a sequence of uniform random numbers, say between –1 and 1, and then transform them using the WHT.

Similarly, you can disrupt the orderly waveform patterns of the WHT by choosing a fixed randomly chosen pattern of sign flips to apply to any input to the transform.  That is equivalent to multiplying the WHT matrix H with a diagonal matrix D of randomly chosen +1, -1 entries giving HD.  The disrupted waveform patterns in HD then fail to correlate with any of the patterns seen in natural data.  As a result, the output of HD has the Normal distribution and is actually a fast Random Projection of the natural data.  Random projections have a wide number of applications in machine learning, such as locality sensitive hashing, compressive sensing, random projection trees, neural network pre and post-processing etc.

References

Walsh (Hadamard) Transform:

Normal Distribution:

Random Projections:

Other Applications:

Related:


PlatoAi. Web3 Reimagined. Data Intelligence Amplified.
Click here to access.

Source: https://www.kdnuggets.com/2021/07/wht-simpler-fast-fourier-transform-fft.html

Continue Reading

Big Data

Must-Know Text Operations in Python before you dive into NLP!

Published

on



Text Operations in Python | Must-Know Text Operations in Python for NLP





















Learn everything about Analytics



PlatoAi. Web3 Reimagined. Data Intelligence Amplified.
Click here to access.

Source: https://www.analyticsvidhya.com/blog/2021/07/must-know-text-operations-in-python-before-you-dive-into-nlp/

Continue Reading

Big Data

Canada’s Rogers Communications beats quarterly revenue estimates

Published

on

(Reuters) -Canada’s Rogers Communications Inc on Wednesday reported second-quarter revenue that beat analysts’ estimates, helped by a pick up in advertisement sales and as its cable business benefited from a pandemic-driven shift to remote work and entertainment.

The requirement of high-speed broadband networks to carry on remote work helped the telecom operator negate the slow recovery from its wireless business.

The return of live sport broadcasting also played a positive role in boosting the Toronto-based telecom operator’s revenue.

The company’s total revenue rose to C$3.58 billion ($2.82 billion) in the quarter ended June 30, compared with analysts’ average estimates of C$3.56 billion, according to IBES data from Refinitiv.

Earlier in March, Rogers said it would buy Shaw Communications Inc for about C$20 billion ($16.02 billion), aiming to double down on its efforts to roll out 5G throughout the country.

Revenue for its cable unit, which includes internet, phone and cloud-based services, rose 5% during the quarter

Quarterly net income rose to C$302 million, or 60 Canadian cents per share, from C$279 million, or 54 Canadian cents, a year earlier.

($1 = 1.2686 Canadian dollars)

(Reporting by Tiyashi Datta in Bengaluru; Editing by Shailesh Kuber)

Image Credit: Reuters

PlatoAi. Web3 Reimagined. Data Intelligence Amplified.
Click here to access.

Source: https://datafloq.com/read/canadas-rogers-communications-beats-quarterly-revenue-estimates/16522

Continue Reading

Big Data

Climate friendly cooling tech firm gets $50 million from Goldman Sachs

Published

on

By Jane Lanhee Lee

(Reuters) – Chemicals used in air conditioning, freezers and refrigeration have long hurt the environment by destroying the ozone layer and polluting water sources, but technology is starting to change the way we keep cool.

Phononic, a startup based in Durham North Carolina using a material called bismuth telluride to make so-called cooling chips, on Wednesday said it raised $50 million from Goldman Sachs Asset Management.

When electricity runs through the chip the current takes heat with it leaving one side of the chip to cool and the other to heat up, said Tony Atti, Phononic co-founder and CEO.

The chips can be as small as a fraction of a fingernail or as big as a fist depending on how much coolants are needed and have been used to create compact freezers for vaccine transportation or for ice-cream at convenience stores like Circle K, he said. A more recent and fast growing use is to prevent overheating in lidars, laser-based sensors in autonomous cars, and optical transceivers for 5G data transmission, said Atti.

“The historical refrigerants that had been used for vapor compression systems, they are both toxic and global warming contributors,” said Atti. While the global warming impact had been reduced, refrigerants still had issues with toxicity and flammability.

Atti said while the bismuth telluride powder itself is toxic, when it is processed into a semiconductor wafer and made into a chip, it is “benign” and can be recycled or disposed as its meets all chip safety and disposal standards.

The cooling chips are manufactured in Phononic’s own factory in Durham and for mass production the company is working with Thailand based Fabrinet. The freezers for vaccines and ice-cream are built in China by contract manufacturers and carry the brands of Phononic’s customers or in some cases are co-branded, he said.

The funding will be used to build out high-volume manufacturing and to expand Phononic’s markets and product line.

Atti declined to share the latest valuation of Phononic but said it was “north of half a billion dollars”. Previous investors include Temasek Holdings and private equity and venture capital firm Oak Investment Partners. 

(Reporting By Jane Lanhee Lee; editing by Richard Pullin)

Image Credit: Reuters

PlatoAi. Web3 Reimagined. Data Intelligence Amplified.
Click here to access.

Source: https://datafloq.com/read/climate-friendly-cooling-tech-firm-gets-50-million-goldman-sachs/16521

Continue Reading
Esports3 days ago

How to reduce lag and increase FPS in Pokémon Unite

Esports4 days ago

Coven skins for Ashe, Evelynn, Ahri, Malphite, Warwick, Cassiopeia revealed for League of Legends

Esports4 days ago

Will New World closed beta progress carry over to the game’s full release?

Aviation5 days ago

And Here’s Yet Another Image Of Russia’s New Fighter Concept That Will Be Officially Unveiled Tomorrow

Esports4 days ago

Can you sprint in New World?

Esports3 days ago

How to add friends and party up in New World

Esports3 days ago

How to claim New World Twitch drops

AR/VR3 days ago

Moth+Flame partners with US Air Force to launch Virtual Reality sexual assault prevention and response training

Esports5 days ago

How to complete FUTTIES Alessandrini’s objectives in FIFA 21 Ultimate Team

Esports3 days ago

Twitch streamer gets banned in New World after milking cow

Esports5 days ago

Everything we know about Seer in Apex Legends

Aerospace5 days ago

Boeing crew capsule mounted on Atlas 5 rocket for unpiloted test flight

Esports4 days ago

What Time Does League of Legends Patch 11.15 Go Live?

Esports5 days ago

Evil Geniuses top laner Impact breaks all-time LCS early-game gold record in win over Dignitas

Blockchain4 days ago

Rothschild Investment Purchases Grayscale Bitcoin and Ethereum Trusts Shares

Blockchain4 days ago

Uniswap (UNI) and AAVE Technical Analysis: What to Expect?

Esports4 days ago

Konami unveils Yu-Gi-Oh! Master Duel, a digital version of the Yu-Gi-Oh! TCG and OCG formats

Blockchain3 days ago

BNY Mellon Joins State Street Into Crypto Trading, Backs Pure Digital Trading Platform

Esports3 days ago

How to change or join a new world in New World

Esports4 days ago

Team BDS adds GatsH to VALORANT roster as sixth man before EU Stage 3 Challengers 2

Trending