DeepMind and ML Graphcast Forecasts Weather

Interesting outcome ....  Similar in other kinds of forecasting?

DeepMind & Google’s ML-Based GraphCast Outperforms the World’s Best Medium-Range Weather Forecasting System

Medium-range weather forecasts play a crucial role in agriculture, construction, travel and other industries. They also bring practical value to people’s daily lives, enabling us to plan outings and keeping us safe from extreme weather events. Traditional numerical weather prediction (NWP)-based forecasting models that run simulations on computing clusters however do not scale efficiently with today’s increasing weather data availability, and their accuracy relies on manual input from experts, which is time-consuming and cost inefficient.

In the new paper GraphCast: Learning Skillful Medium-Range Global Weather Forecasting, a research team from DeepMind and Google presents GraphCast, a machine-learning (ML)-based weather simulator that scales well with data and can generate a 10-day forecast in under 60 seconds. GraphCast outperforms the world’s most accurate deterministic operational medium-range weather forecasting system and all existing ML-based benchmarks.

The team summarizes their work’s key advances as follows:

A novel multi-mesh GNN architecture for learned weather simulation.

An autoregressive model that can be trained to generate forecasts on 0.25° latitude-longitude resolution and 37 levels of vertical resolution, for 40 or more steps.

An evaluation protocol with comprehensive coverage of medium-range forecast variables.

An ML-based forecasting model with greater skill than the best NWP-based deterministic model.

The most accurate ML-based weather forecasting model. ... ' 

Data and Demand Forecasting

Converting retail data to demand forecasting. 

Snowflake readies Amazon Retail Data for Demand Forecasting

Andrew Brust  in Venturebeat   March 29, 2022 6:00 AM

To further strengthen our commitment to providing industry-leading coverage of data technology, VentureBeat is excited to welcome Andrew Brust and Tony Baer as regular contributors. Watch for their articles in the Data Pipeline.

Cloud data warehouse darling Snowflake has been investing a lot in its data ecosystem. Yesterday, the company made a broad announcement around the launch of its Retail Data Cloud, which integrates the core Snowflake platform with partner-delivered solutions and industry-specific datasets. Today, the company is announcing an important initiative within that rollout, which it shared exclusively with VentureBeat: it’s partnering with Amazon Web Services (AWS) to bring Amazon.com sales channel data directly into customers’ Snowflake data warehouse instances.

Data for sales intelligence

This is a data-driven age for retail, a phenomenon accelerated by the pandemic and the twists and turns COVID has created in buyer preferences and purchasing patterns. The provision of the Amazon.com data is intended to help retail and CPG (consumer packaged goods) customers not only monitor Amazon.com Vendor Central PO data, but also leverage Amazon Forecast functionality within their Snowflake environments.  ... "

AI-driven Operations Forecasting in Data-light Environments

 Integrating accurate forecasting is essential.

AI-driven operations forecasting in data-light environments

For better forecasting in operations management, AI is proving essential. And limited data is no longer the barrier it once was.

Sent from McKinsey Insights, available in the App Store and Play Store.

AI-driven operations forecasting in data-light environments    February 15, 2022  

By Jorge Amar, Sohrab Rahimi, Zachary Surak, and Nicolai von Bismarck

Too many companies still rely on manual forecasting because they think AI requires better-quality data than they have available. Nowadays, that’s a costly mistake.

What do internal functions as diverse as risk assessment, capital-expenditure planning, and workforce planning have in common? Each is fundamentally about understanding demand—making demand forecasting an essential analytical process. Amid rising pressure to increase forecasting accuracy, more companies have come to rely on AI algorithms, which have become increasingly sophisticated in learning from historical patterns.

AI models have clear advantages over traditional spreadsheet-based analytic methods. Applying AI-driven forecasting to supply chain management, for example, can reduce errors by between 20 and 50 percent—and translate into a reduction in lost sales and product unavailability of up to 65 percent. Continuing the virtuous circle, warehousing costs can fall by 5 to 10 percent, and administration costs by 25 to 40 percent. Companies in the telecommunications, electric power, natural gas, and healthcare industries have found that AI forecasting engines can automate up to 50 percent of workforce-management tasks, leading to cost reductions of 10 to 15 percent while gradually improving hiring decisions—and operational resilience  (Exhibit 1).  .... 

Enabling AI-driven forecast models

We strive to provide individuals with disabilities equal access to our website. If you would like information about this content we will be happy to work with you. Please email us at: McKinsey_Website_Accessibility@mckinsey.com

Automated AI-driven forecasting promotes these benefits by consuming real-time data and continuously identifying new patterns. This capacity enables fast, agile actions because the model anticipates demand changes rather than just responding to them. In contrast, traditional approaches to demand forecasting require constant manual updating of data and adjustments to forecast outputs. These interventions are typically time-consuming and do not allow for agile responses to immediate changes in demand patterns.  ....' 

NVIDIA Time Series Forecasting

And time series forecasting is a frequently useful need in anaytics.  NVIDIA provides a means, have not examined closely to see how complete this is. 

 Time Series Forecasting with the NVIDIA Time Series Prediction Platform and Triton Inference Server

By Kyle Kranen, Pawel Morkisz and Carl (Izzy) Putterman.  NVIDIA

n this post, we detail the recently released NVIDIA Time Series Prediction Platform (TSPP), a tool designed to compare easily and experiment with arbitrary combinations of forecasting models, time-series datasets, and other configurations. The TSPP also provides functionality to explore the hyperparameter search space, run accelerated model training using distributed training and Automatic Mixed Precision (AMP), and deploy and run inference on accelerated model formats on the NVIDIA Triton Inference Server.

Accurately forecasting future time series values using previous values has proven pivotal in understanding and managing complex systems, including but not limited to power grids, supply chains, and financial markets. In these forecasting applications, single-digit percentage improvements in predictive accuracy can have vast financial, ecological, and social impacts. In addition to needing to be accurate, forecasting models also must be able to function on real-time timescale.  ... '

Cellphone COVID Forecasting

Another example from Cellphone data, had seen previous examples that were unsuccessful 

COVID Forecasting Method Using Hospital, Cellphone Data Can Reliably Guide U.S. Cities Through Pandemic Threats

University of Texas at Austin News, February 2, 2022

Researchers at the University of Texas at Austin (UT) developed a forecasting model and two public-facing dashboards that use cellphone mobility data and COVID-19 hospital admissions data to forecast regional hospital demands. The models and dashboards have been used for nearly two years to manage healthcare resources, maintain sufficient hospital capacity, and inform the public about pandemic risks. Using intuitive graphics and spaghetti lines from hurricane forecasting (computer models that show potential tropical cyclone paths), the dashboards display immediate and future risks of COVID-19’s spread. The model, which uses anonymized cellphone mobility data from SafeGraph, can be adapted to project COVID-19 healthcare needs three weeks in advance for any U.S. city. UT's Spencer Fox said, "Community movement data helps us gauge changing COVID-19 transmission risks and anticipate healthcare surges several weeks in advance."  .... 

IBM Weather Company Best Weather Forecast

 We actively used weather prediction data and programs for supply chain models.  I now use IBM's Weather Company almost daily. Quite good, very accurate, not perfect.   Often small errors can make a big difference.   Risk for these small errors need to be considered.  Forecasting is still hard unless you can find and accurately quantify all inputs.  Forecasting is usually only one of a number of things that need to be forecast to optimize supply chains.

IBM's Weather Company the Most Accurate Forecaster Overall  in PRNewswire.

NEW YORK and ATLANTA, July 29, 2021 /PRNewswire/ -- IBM (NYSE: IBM) and its subsidiary The Weather Company, which includes The Weather Channel digital properties, were determined to be "the overall most accurate provider globally" by ForecastWatch, a premier organization for evaluating the accuracy of weather forecasts. In its latest comprehensive study of forecast accuracy released today,ii Global and Regional Weather Forecast Accuracy Overview, 2017-2020, commissioned by IBM, ForecastWatch   ... '

Superforecasting Update

For some time have followed GoodJudgment  and their 'superforecasting' services.   Most recently got a new overview with considerable efforts and claims.    Since very accurate forecasting would seem to be fundamentally unbeatable in many domains, why believe it?  May depend on the definition of a correct forecast in a given context.  See their work with returns to normality with Covid.  And 'Right ways to think about the future'.

Why Superforecasting?

When you make decisions based on precise probability forecasts, rather than hunches, the benefits are game-changing.

Gain a competitive edge

Making decisions without precise probability forecasts is like playing poker without counting cards. Outsmart your competition with well-calibrated forecasts.

Manage risk

Convert strategic uncertainty into manageable risk to avoid costly mistakes.

Seize opportunities

Identify expected outcomes before others recognize them so that you can capitalize on opportunities quickly and with confidence.  From groundbreaking theory to powerhouse practice

In 2011, IARPA – the US intelligence community’s equivalent to DARPA – launched a massive competition to identify cutting-edge methods to forecast geopolitical events. Four years, 500 questions, and over a million forecasts later, the Good Judgment Project (GJP) – led by Philip Tetlock and Barbara Mellers at the University of Pennsylvania – emerged as the undisputed victor in the tournament. GJP’s forecasts were so accurate that they even outperformed intelligence analysts with access to classified data.

Good Judgment Inc is now making this winning approach to harnessing the wisdom of the crowd available for commercial use. Our clients benefit from the externally validated forecasting methodology that made the Good Judgment Project so successful.

Today, Good Judgment’s professional Superforecasters deliver unparalleled accuracy on forecasting questions across the political, economic and social spectrum. And, we train others to apply this evidence-based methodology within their own teams.  .... '

Using AI for the Supply Chain

 As a long time practitioner in the supply chain space,  here an example.  Notable here the forecast of demand as we transition to new context.  No direct how mentioned here, but taking a look.    

In ACM Tech News

Supply Chain Strains Sharpen Focus on AI   By The Wall Street Journal, April 8, 2021

Supply chains have taken a battering this year from the coronavirus pandemic and other extreme events, and artificial intelligence has emerged as a critical tool for navigating business in this environment.

Artificial intelligence (AI) increasingly is being used to forecast demand more accurately and optimize supply chains.

XPO Logistics Inc.'s Mario Harik said the recent Suez Canal blockage underscores the benefits of AI in quickly rerouting shipments and building redundancy into operations through multiple distribution facilities in preparation for extreme events.

Harik said, "We collect millions of data points from our operation and then we use our machine learning algorithms to be able to help forecast either where demand is going to come from, or how to best optimize a certain part of the supply chain to be as effective as possible."

Forecasting models have gained in importance as companies ramp up production due to the pandemic.

Koninklijke Philips NV's Ivanka Janssen said that with AI, "We see that we get a different level of forecasting accuracy that has a true benefit throughout the supply chain."

From The Wall Street Journal

How AI Will Change Marketing

The point is made that predictive, forecasting and other forward looking capabilities and alerting will be the key to improving marketing.   Those who know more about the future can leverage it.

How artificial intelligence will change be the future of marketing in DSC

Posted by Elice Max on January 16, 2021 at 12:00a 

Artificial intelligence is reshaping almost all the aspects of doing business, from finance to sales and R&D to operations. Its most profound impact is, however, being witnessed in the field of marketing where not only it has already created great value but experts predict it is going to change the future of marketing massively. In this article we will explore how this branch of knowledge that is based on cognitive technologies and replication of human intelligence in machines is going to transform the marketing strategies.

The transformation of marketing by AI has already begun. Using algorithms, handling an extremely huge amount of data, and excellent analytical techniques, the businesses are now adopting customer-centric marketing strategies to communicate with the targeted and segmented audience in an intelligent manner. The fashion industry is a case in point where marketing today is immersive, interactive, and is thus providing customers a cutting-edge shopping industry.

How will AI transform marketing in future?

The marketing transformation is long overdue as it is still using the old playbook using communication with all. The AI’s machine learning is however bound to humanize marketing. Here is what changes it will create in the future.

Marketers will be aided by Strong predictive ability: Using conversational assistance, marketers will be able to dig deep into customers’ needs and ongoing trends and therefore will be capable to predict future trends more accurately.... ' 

Find the Warning Signals

A classic kind of problem we addressed in practice: can we find signals of, or predictive signals of future occurrence?   Powerful idea. How possible is this, very generally in streams of data?  Connection to risk analysis?

Method Finds Hidden Warning Signals in Measurements Collected Over Time

MIT News  Daniel Ackerman

A new technique developed by researchers at the Massachusetts Institute of Technology (MIT) and Spain's Rey Juan Carlos University uses a generative adversarial network (GAN) to identify anomalies in time series data. The TadGAN approach could be used to detect and respond to significant changes in a range of high-value systems. The researchers aimed to create a general framework for anomaly detection that could be applied across industries. TadGAN can distinguish between normal and anomalous data points by checking for discrepancies between the real-time series and a fake GAN-generated time series. Supplementing the GAN with an autoencoder algorithm helped the researchers strike a balance between being vigilant and raising too many false alarms. In anomaly detection tests on 11 datasets, TadGAN outperformed ARIMA, the traditional approach to time-series forecasting, for eight datasets. Said MIT's Sarah Alnegheimish, "We want to mitigate the stigma about artificial intelligence not being reproducible."

Timing is the Thing for Modeling the Risk

Forecasting once again is essential for knowing how to react.

Supply chain outlook: The timing of the slowdown
MIT Professor David Simchi-Levi forecast the mid-March manufacturing pause. Now he looks ahead.

Peter Dizikes | MIT News Office
March 25, 2020

With the Covid-19 virus disrupting the global economy, what is the effect on the international supply chain? In a pair of articles, MIT News examines core supply-chain issues, in terms of affected industries and the timing of unfolding interruptions.

The rapid spread of the Covid-19 virus is already having a huge impact on the global economy, which is rippling around the world via the long supply chains of major industries.

MIT supply chain expert David Simchi-Levi has been watching those ripples closely in 2020, as they have moved from China outward to the U.S. and Europe. His tracking of supply chain problems provides insight into what is happening in the global economy — and what could happen in a variety of scenarios.

“This is a significant challenge,” says Simchi-Levi, who is a professor of engineering systems in the School of Engineering and in the Institute for Data, Systems, and Society within the MIT Stephen A. Schwarzman College of Computing. The global public health crisis, he adds, “is not only affecting the supply chain. There is a significant impact on demand, and as a result, a significant impact on the financial performance on all these businesses.”   ... "

Superforecasting Short Course from The Edge

Quite interesting, free, have followed the concept for some time, but have yet to apply it.  Videos embedded at the link.   Good to see this followup.  Is enough said about embedded risk analysis?  Following.

A Short Course in Superforecasting—Philip Tetlock: An EDGE Master Class
(ED. NOTE: In 2015, Edge presented "A Short Course in Superforecasting" with political and social scientist Philip Tetlock. Superforecasting is back in the news this week thanks to the UK news coverage of comments by Boris Johnson's chief adviser Dominic Cummings, who urged journalists to "read Philip Tetlock's Superforecasters [sic], instead of political pundits who don't know what they're talking about.")

PHILIP E. TETLOCK, political and social scientist, is the Annenberg University Professor at the University of Pennsylvania, with appointments in Wharton, psychology and political science. He is co-leader of the Good Judgment Project, a multi-year forecasting study, author of Expert Political Judgment, co-author of Counterfactual Thought Experiments in World Politics (with Aaron Belkin), and co-author of Superforecasting: The Art & Science of Prediction (with Dan Gardner). Further reading on Edge: "How to Win at Forecasting: A Conversation with Philip Tetlock" (December 6, 2012). Philip Tetlock's Edge Bio Page.

CLASS I — Forecasting Tournaments: What We Discover When We Start Scoring Accuracy
It is as though high status pundits have learned a valuable survival skill, and that survival skill is they've mastered the art of appearing to go out on a limb without actually going out on a limb. They say dramatic things but there are vague verbiage quantifiers connected to the dramatic things. It sounds as though they're saying something very compelling and riveting. There's a scenario that's been conjured up in your mind of something either very good or very bad. It's vivid, easily imaginable.

It turns out, on close inspection they're not really saying that's going to happen. They're not specifying the conditions, or a time frame, or likelihood, so there's no way of assessing accuracy. You could say these pundits are just doing what a rational pundit would do because they know that they live in a somewhat stochastic world. They know that it's a world that frequently is going to throw off surprises at them, so to maintain their credibility with their community of co-believers they need to be vague. It's an essential survival skill. There is some considerable truth to that, and forecasting tournaments are a very different way of proceeding. Forecasting tournaments require people to attach explicit probabilities to well-defined outcomes in well-defined time frames so you can keep score.   ... " 

MIT Model Forecasts Business Financials

If real, here is something of considerable value ... Note 'Belief Propagation' .... which we tested for things like predicting future product sales in varying context.  Looking more closely.

Model Beats Wall Street Analysts in Forecasting Business Financials
MIT News
By Rob Matheson

Massachusetts Institute of Technology (MIT) researchers have developed an automated model that significantly outperforms humans in predicting business sales using only anonymized weekly credit card transactions and three-month earnings reports. The researchers used the model to predict quarterly earnings of more than 30 companies, and found that it outperformed the combined estimates of expert Wall Street analysts on 57% of predictions. The human analysts had access to any available private or public data and other machine learning models, while the MIT model used a very small dataset of the two data types. The researchers used a variation of the standard inference algorithm, called Kalman filtering or Belief Propagation. This technique uses data measurements observed over time, containing noise inaccuracies, to generate a probability distribution for unknown variables over a designated timeframe .... " 

Towards Better Forecasting: Less Noise

A forecast prediction is key for any business decision. Short intro, then the entire podcast follows at the link.

Wharton’s Barbara Mellers and Ville Satopӓӓ from INSEAD discuss their research on the impact of noise on forecast accuracy.

From predicting the weather to possible election outcomes, forecasts have a wide range of applications. Research shows that many forces can interfere with the process of predicting outcomes accurately — among them are bias, information and noise. Barbara Mellers, a Wharton marketing professor and Penn Integrates Knowledge (PIK) professor at the University of Pennsylvania, and Ville Satopӓӓ, assistant professor of technology and operations management at INSEAD, examined these forces and found that noise was a much bigger factor than expected in the accuracy of predictions. The professors recently spoke with Knowledge@Wharton about their working paper, “Bias, Information, Noise: The BIN Model of Forecasting.” (Listen to the podcast at the top of this page.)

An edited transcript of the conversation follows.

Knowledge@Wharton: In your paper, you propose a model for determining why some forecasters and forecasting methods do better than others. You call it the BIN model, which stands for bias, information and noise. Can you explain how these three elements affect predictions?

Ville Satopӓӓ: Let me begin with information. This describes how much we know about the event that we’re predicting. In general, the more we know about it, the more accurately we can forecast. For instance, suppose someone asked me to predict the occurrence of a series of future political events. If I’m entirely ignorant about this, I don’t really follow politics, I don’t follow the news, I barely understand the questions you’re asking me, I would predict around 50% for these events.

On the other hand, suppose I follow the news and I’m interested in the topic. My predictions would be then more informed, and hence they would be not around 50% anymore. Instead, they would start to tilt in the direction of what would actually happen.

At the extreme case, we could think of me having some sort of a crystal ball that would allow me to see into the future. This would make me perfectly informed, and hence I would predict zero or 100% for each one of the events, depending on what I see in the crystal ball. This just illustrates how information can drive our predictions. It introduces variability into them that is useful because it is based on actual information. Because of that, it correlates with the outcome.   .... " 

Attention for Advanced Forecasting and Classification

Interesting and quite technical view of forecasting and classification that is worth a look.  Of course accurate and timely forecasting is important for most businesses.  Considerable piece here, below an intro with much more at the link.  Have never seen it done accurately enough with these kinds of methods.

Attention for time series forecasting and classification
Harnessing the most recent advances in NLP for time series forecasting and classification  By Isaac Godfried

Transformers (specifically self-attention) have powered significant recent progress in NLP. They have enabled models like BERT, GPT-2, and XLNet to form powerful language models that can be used to generate text, translate text, answer questions, classify documents, summarize text, and much more. With their recent success in NLP one would expect widespread adaptation to problems like time series forecasting and classification. After all, both involve processing sequential data. However, to this point research on their adaptation to time series problems has remained limited. Moreover, while some results are promising, others remain more mixed. In this article, I will review current literature on applying transformers as well as attention more broadly to time series problems, discuss the current barriers/limitations, and brainstorm possible solutions to (hopefully) enable these models to achieve the same level success as in NLP. This article will assume that you have a basic understanding of soft-attention, self-attention, and transformer architecture. If you don’t please read one of the linked articles. You can also watch my video from the PyData Orono presentation night.

Attention for time series data: Review

The need to accurately forecast and classify time series data spans across just about every industry and long predates machine learning. For instance, in hospitals you may want to triage patients with the highest mortality early-on and forecast patient length of stay; in retail you may want to predict demand and forecast sales; utility companies want to forecast power usage, etc. .... " 

Autonomous Planning in Food Retail

Advanced Planning using machine learning techniques are suggested for procurement.   We always had advanced planning, its just how well you could integrate it with predictions of demand, and in particular unusual elements of forecasts.   New analytics have emerged, but how well will they deliver?

The invisible hand: On the path to autonomous planning in food retail   from McKinsey

It’s not news to food retailers: sometimes your stocks are too high, sometimes they’re too low. Advanced planning now gives them entirely new options for solving the expensive problem—and cuts costs in the process.  .... 

Procurement planners in food retail today are not to be envied. They have to please customers who have never made more exacting demands on availability, freshness, and range. And they ignore such expectations at their peril: the competition is relentless, driving all market participants to seek out improvements incessantly. Those who stick to their legacy processes can only make comparable progress at the cost of mounting stocks, increasing write-offs, and an increasingly complex supply chain.

Internally, planners are often struggling with outdated IT systems that are isolated from each other, unreliable sources of information, and in some cases, largely manual and poorly coordinated processes. Forecasts are commensurately inaccurate and personnel expenses high. Externally, on the other hand, decision makers are faced with an increasingly unfathomable offering from digital service providers that—although they can process huge volumes of data with their solutions—cannot give retailers any advantages of relevance as long as they leave their operating models unchanged.


The future will likely be very different. A look at online retail already reveals the shape of things to come: leading companies are developing highly integrated planning systems that already use the most advanced analytics and machine-learning solutions available today. These high-tech methods, also referred to as “advanced planning,” will, in the future, take control of steering in food retail as well. And they set exacting requirements on companies: they entail tapping the entire wealth of transaction data along with external parameters as sources. Retailers need a completely different process landscape, new capabilities, more computing power, and advanced algorithms.  ... " 

Predicting Severe Weather

I like this experiment because there is so much data gathered and involved,  and many current models, so it should be easy to do good comparisons of methods. I note that the emphasis may be on short term forecasting.

Machine learning and its radical application to severe weather prediction  by Eric Verbeten, University of Wisconsin-Madison

In the last decade, artificial intelligence ("AI") applications have exploded across various research sectors, including computer vision, communications and medicine. Now, the rapidly developing technology is making its mark in weather prediction.

The fields of atmospheric science and satellite meteorology are ideally suited for the task, offering a rich training ground capable of feeding an AI system's endless appetite for data. Anthony Wimmers is a scientist with the University of Wisconsin–Madison Cooperative Institute for Meteorological Satellite Studies (CIMSS) who has been working with AI systems for the last three years. His latest research investigates how an AI model can help improve short-term forecasting (or "nowcasting") of hurricanes.

Known as DeepMicroNet, the model uses deep learning, a type of neural network arranged in "deep" interacting layers that finds patterns within a dataset. Wimmers explores how an AI system like DeepMicroNet can supplement and support conventional weather prediction systems.  .... " 

Gartner Blog: We are Very early in Predictive Analytics

Here an excerpt.    Interesting thought, and our forecasting will become much better, in some areas spectacularly so.  But still, it will be operating with some of the same data. Old, faulty, in the wrong context, gathered haphazardly.  So you still cannot expect exact predictions.  If you are not exactly right there is the risk of implemented error.  And that risk itself cannot be perfectly tagged.   So there again a caution.   And I would certainly not use the position of a technology on a wavy line as predictive input data.  So please, save your pennies, but caution all around.

Start Saving for Predictive Analytics
by Steve Rietberg  |  August 8, 2019  | 

When I was young and fresh out of college, people urged me to start contributing to a retirement fund as soon as I could. They explained that since I had time on my side, even modest investments would pay significant dividends. I just needed the discipline to start saving early.

We’re all young, with respect to predictive analytics technology. According to Gartner’s Hype Cycle for CRM Sales, sales predictive analytics (which includes predictive forecasting, upsell/cross-sell recommendations and opportunity scoring) is still in its adolescence. This market is expected to grow quickly in the immediate future.

How can we take advantage of our youth, and get a head start on preparing for advances in predictive pipeline analytics?

Your CRM Thinks Too Linearly

Consider this. Your historic pipeline data suggests a correlation between sales stage and an opportunity’s likelihood to close. In many organizations, that correlation–directly or indirectly–informs seller coaching and forecasting decisions. But there is a disconnect between the sequential sales stages in a typical CRM system and the nonlinear buying process that customers actually follow. To improve the validity and accuracy of your pipeline analytics, you need to track more than sales stage. For this reason, measuring buyer behavior along with seller-provided sales stage and probability clears a path to improved predictive analytics. .... " 

Deep Learning for Time Series and Signal Processing

We researched the topic of using deep learning to replace forecasting time series applications in the enterprise,  for faster and more accurate results, but were never successful.    The need to segment many small  time steps and solve the resulting large neural nets was too difficult.   Here a useful piece on the current related approaches with machine learning.   Reading and will comment further.  From DSC.   Join the group

Deep learning: the final frontier for signal processing and time series analysis?
Posted by Andrea Manero-Bastin  
This article was written by Alexandr Honchar.

People use deep learning almost for everything today, and the “sexiest” areas of applications are computer vision, natural language processing, speech and audio analysis, recommender systems and predictive analytics. But there is also one field that is unfairly forgotten in terms of machine learning — signal processing (and, of course, time series analysis). In this article, I want to show several areas where signals or time series are vital, after I will briefly review classical approaches and will move on to my experience with applying deep learning for biosignal analysis in Mawi Solutions and for algorithmic trading. I already gave a couple of talks on this topic in Barcelona and Lviv, but I would like to make the materials a bit more accessible.

I am sure, that not only people working with time series data will benefit from this article. Computer vision specialists will learn how similar their domain expertise is to signal processing, NLP people will get some insights about sequential modeling and other professionals can have interesting takeaways as well. Enjoy!  ... '

AI Impact on Demand Forecasting

Good short piece,  makes good points about what is needed, and should be expected.  I have taught forecasting in the enterprise, and its more about how the forecast is used than what it is.   You would love to have it perfect, but it will not be.   AI provides another useful component.

Is AI’s impact on demand forecasting more hype than reality?   by Nikki Baird in Retailwire

Through a special arrangement, presented here for discussion is a summary of a current article from the blog of Nikki Baird, VP of retail innovation at Aptos. The article first appeared on Forbes.com.

The forecast error in retail is as high as 32 percent, according to some estimates. Will artificial intelligence (AI) technology do any better?

AI promises to change the way demand forecasting works in retail in six key ways, but those promises include a bit of hype:  .... "