Hexbyte  Hacker News  Computers Data Mining Reveals the Crucial Factors That Determine When People Make Blunders

Hexbyte Hacker News Computers Data Mining Reveals the Crucial Factors That Determine When People Make Blunders

Hexbyte Hacker News Computers

The way people make decisions in the real world is a topic of increasing interest among psychologists, social scientists, economists, and others. It determines how economies perform, how elections are run, and how conflicts break out and get resolved.

One idea has provided a focal point for decision-making research. This is the notion of bounded rationality—that people are limited by various constraints in the real world, and these play a crucial role in the decision-making process.  People are limited by the difficulty of the decision they have to make, their own decision-making skill, and the time they can spend on the problem. Nevertheless, whatever the circumstances, a decision has to be made and the consequences accepted.

That raises an important set of questions. How do these factors influence the quality of the decision being made? Does time pressure have a bigger impact than, say, decision-making skill on the quality of a decision?

These are hard questions to answer, given the difficulty of setting up a controlled experiment to test them. Indeed, nobody has found a satisfactory way of studying the problem.

Until now. Today, Ashton Anderson at Microsoft Research in New York City, Jon Kleinberg at Cornell University in Ithaca, and Sendhil Mullainathan at Harvard University in Cambridge unveil the first large-scale study of decision making under controlled conditions. For the first time, these guys have been able to study how the quality of decision making changes with the time available, the skill of the decision maker, and the difficulty of the decision at hand.

Their laboratory? The game of chess. “We have used chess as a model system to investigate the types of features that help in analyzing and predicting error in human decision-making,” they say.

Their research focuses on a database of 200 million chess games played online between amateurs and another database of around one million games played between grand masters. What’s interesting about these databases is that the outcome of the game reveals whether a player has made a mistake. And the recorded moves reveal exactly when the losing player makes the blunder.

The team are then able to see what factors have played a role. They can see whether the player was under time pressure, for example. They can see the difficulty of the decision by examining the position on the board and its complexity. They do this by totting up all possible moves and then working out what fraction of them are blunders. So a position in which all moves except one are blunders is more difficult than a position in which only one move out of many is a blunder.

The team also know the skill level of the players. The skill level of every chess player is given by a number called the Elo rating (after Arpad Elo, who came up with it). Most amateurs have a rating between 1000 and 2000, strong amateurs get up to 2400, and the world’s top players receive rankings of around 2600. There are generally just a handful of players at any time with a ranking over 2800. A difference of 400 points between players suggests that the higher-ranked player is overwhelmingly likely to win.

And the huge size of the database allows them to cut and dice the data in a way that holds two of these variables constant while allowing the other to vary. For example, the team can examine board positions of the same difficulty while players are under the same time pressure to see how any variation in their skill level influences the quality of the decisions they make. Equally, the researchers can hold skill and time pressure constant while allowing board position to vary; and so on.

The results make for interesting reading. They find, for example, that the amount of time spent on a decision is a factor in blundering, but only up to a point. Quick decisions are more likely to lead to a blunder, but after about 10 seconds or so the likelihood of a blunder flattens out. So when players spend more time than this on a move, it is probably because they don’t know what to do.

The difficulty of the decision is an important factor, too. More difficult positions are more likely to lead to a blunder. And skill levels have a big impact in reducing the likelihood of a blunder. In general, better players make better decisions.

But Anderson and co have found evidence of an entirely counterintuitive phenomenon in which skill levels play the opposite role, so that skillful players are more likely to make an error than their lower-ranked counterparts. The team call these “skill anomalous positions.”

That’s an extraordinary discovery which will need some teasing apart in future work. “The existence of skill-anomalous positions is surprising, since there is a no a priori reason to believe that chess as a domain should contain common situations in which stronger players make more errors than weaker players,” say Anderson and co. Just why this happens isn’t clear.

These results have an important application. They allow the team to predict when a player is most likely to make a mistake. And it turns out that one of the factors is a much more powerful predictor than the others.

The bottom line is that the difficulty of the decision is the most important factor in determining whether a player makes a mistake.  In other words, examining the complexity of the board position is a much better predictor of whether a player is likely to blunder than his or her skill level or the amount of time left in the game.

That could have important implications for the way researchers examine other decisions. For example, how does the error rate of highly skilled drivers in difficult conditions compare with that of bad drivers in safe conditions? If the difficulty of the decision is the crucial factor, rather than driver skill, then much more emphasis needs to be placed on this. “We think of inexperienced and distracted drivers as a major source of risk, but how do these effects compare to the presence of dangerous road conditions?” ask Anderson and co.

And given the team’s discovery of skill-anomalous conditions, are there road conditions that make skillful drivers more likely to make a mistake than less skillful ones?

This kind of work will have big implications beyond driving. Economists might well ask what all this means for buying decisions, election officials will ask about the complexity of information related to voting decisions, and negotiators will think about its impact on resolving conflict.

Fascinating work and plenty of food for thought.

Ref: arxiv.org/abs/1606.04956  : Assessing Human Error Against a Benchmark of Perfection

Read More

Hexbyte  Hacker News  Computers Crypto mining giant Bitmain reveals heady growth as it files for IPO

Hexbyte Hacker News Computers Crypto mining giant Bitmain reveals heady growth as it files for IPO

Hexbyte Hacker News Computers

After months of speculation, Bitmain — the world’s largest provider of crypto miners — has opened the inner details of its business after it submitted its IPO prospectus with the Stock Exchange of Hong Kong. And some of the growth numbers are insane.

The document doesn’t specify how much five-year-old Bitmain is aiming to raise from its listing — that’ll come later — but it does lift the lid on the incredible business growth that the company saw as the crypto market grew massively in 2017. Although that also comes with a question: can that growth continue in this current bear market?

The company grossed more than $2.5 billion in revenue last year, a near-10X leap on the $278 million it claims for 2016. Already, it said revenue for the first six months of this year surpassed $2.8 billion.

Bitmain is best known for its ‘Antminer’ devices — which allow the owner to mine for Bitcoin and other cryptocurrencies — and that accounts for most of its revenue: 77 percent in 2016, 90 percent in 2017, and 94 percent in the first half of 2018. Other income is generated by its mining farms, shared mining pools, AI chips and blockchain services.

The company is fabless, which means it develops its own chip design and works with manufacturing partners who bring them to life as physical chips. Those chips are then used to power mining hardware which lets the owner earn a reward by mining Bitcoin and other cryptocurrencies. Bitmain claims over 80,000 customers with just under half of sales in China and the rest overseas.

The company said it posted $701 million in net profit in 2017, up from $104 million in 2016. For the first half of this year, it is claiming a gross profit of $743 million. (Operational profit touched $1 billion for that period.)

Hexbyte  Hacker News  Computers

That’s quite staggering growth, but there are some signs that 2018 comes with more challenges.

Margins are down. Gross margin in the first six months was 36 percent, down from 48 percent in 2017 and 54 percent in 2016. Contributing to that, the cost of sale percentage in the first half of 2018 rose to 64 percent from 51 and 52 percent in 2017 and 2016, respectively.

Bitmain is trying to bat away those concerns by using H1 2018 figures, rather than splitting that period into two quarters. That’s important because the crypto market has plunged massively since January, losing more than half of its value. That has impacted most crypto companies — whether it is exchanges seeing less trading or wallets less traffic — and it is sure to have had a toll on Bitmain.

The question is to what extent?

That’s crucial because it is what will give this IPO momentum, but Bitmain isn’t playing ball and showing us the full picture.

Hexbyte  Hacker News  Computers

Interestingly, Bitmain accepts Bitcoin and other cryptocurrencies as payment for its miners, with some 27 percent of purchases last year paid for using crypto. As a result, those payments aren’t included in revenue but do show up as “investing cash inflow” when they are converted to fiat and used in the business. That’s a 2018 accounting problem right there.

As a result, Bitmain has a negative net cash used in operating activities position but those become positive when factoring in the crypto. The company said it held $887 million in crypto as of the end of the first half of 2018, that’s up from $872 million in 2017, $56 million in 2016 and $12 million in 2015. The company said that changes in the market saw it lose $102.7 million in value from its crypto hoard. During the first six months of 2018, it cashed out $516.5 million worth of crypto, having exchanged $529 million in 2017.

The wild ride of 2017, however, led the company to over-estimated demand and, as a result, its inventory ballooned by $1 billion.

Here’s Bitmain explanation of how it managed to get it so wrong:

In early 2018, we anticipated strong market growth for cryptocurrency mining hardware in 2018 due to the upward trend of cryptocurrencies price in the fourth quarter of 2017, and we placed a large amount of orders with our production partners in response to the anticipated significant sales growth. However, there had been significant market volatility in the market price of cryptocurrencies in the first half of 2018. As a result of such volatility, the expected economic return from cryptocurrency mining had been adversely affected and the sales of our mining hardware slowed down, which in turn caused an increase in our inventories level and a decrease in advances received from our customers in the first half of 2018. Going forward, we will actively balance our business growth strategy, inventories and cryptocurrency asset levels to ensure a sustainable business growth and a healthy cash flow position, and we will adjust our procurement and prediction plan to maintain an appropriate liquidity level.

Despite an extra $1 billion in inventory, Bitmain estimates it has the working capital — including crypto pile and the result of its IPO — to sustain operations for at least another 12 months. That, according to its figures, is around $343 million in cash and cash equivalents but clearly it needs another megahit product or for the market demand to rise again.

Indeed, Bitmain just last week announced its newest mining chip — shrunk down to 7nm — which it believes will offer more power and greater efficiency for miners. That progress coupled with the rising value of crypto — i.e. what owners of Bitmain miners can earn — has helped the company steadily raise the price of its hardware.

Average selling price for its Bitcoin mining machines in 2015 was just $463, but that jumped to $767 in 2016, $1,231 in 2017 and $1,012 in the first half of 2018.

Hexbyte  Hacker News  Computers  

Bitmain co-founder Jihan Wu is the face of the company and one of its largest shareholders with a 20 percent stake

Beyond mining, the company is also developing AI chips, the first of which launched last year. They are used for developing cloud systems, as well as object, image and facial recognition purposes.

Citing third party figures, Bitmain claims to have a dominant 75 percent of the ASIC mining hardware market. It is investing heavily in R&D, which reached $73 million last year and $86 million during the first half of 2018. In addition, around one-third of its 2,594 employees are listed as working in research and development.

Hexbyte  Hacker News  Computers

It’s likely that Bitmain sees more revenue in crypto than any other company on the planet

Bitmain’s document confirms the company raised some $784 million across Series A, Series B and Series B rounds.

Its investor roster is fairly public thanks to leaks and it includes the likes of IDG, Sequoia China, and Kaifu Lee’s Sinovation fund. However, the prospectus does confirm that shareholders include retailer NewEgg, EDBI — the corporate investment arm of Singapore’s Economic Development Board — and Uber investor Coatue. Founders Ketuan Zhan and Jihan Wu are the largest shareholders and they control 36 and 20 percent, respectively.

We can expect Bitmain to flesh out the prospectus with more juicy information, including a target raise which will also generate its valuation. But for now there are over 400 pages of information to process, you can find them all right here.

Note: The original version of this article has been updated to correct the figures for Bitmain’s crypto holdings.

Editorial note: The author owns a small amount of cryptocurrency. Enough to gain an understanding, not enough to change a life.

Read More