Minimizing Multimodular Functions and Allocating Capacity in Bike-Sharing Systems

Daniel Freund, Shane G. Henderson, David B. Shmoys (2022) Minimizing Multimodular Functions and Allocating Capacity in Bike-Sharing Systems. Operations Research 70(5):2715-2731. (Best OM Paper in Operations Research Award, arXiv)

如何準備研究所

• 先問自己，真的想念研究所以充實自我？逃避就業？爸媽的要求？補習班的鼓吹？
• 一定要念研究所？大學如果認真念，已經足夠找到理想的工作。如果不確定，可以先工作；工作後覺得不足夠，再報考您需要加強的領域，學習動機也比較高。

Becoming a Data Head

Alex J. Gutman and Jordan Goldmeier, Becoming a Data Head: How to Think, Speak, and Understand Data Science, Statistics, and Machine Learning, Wiley, 2021.

作業研究期末報告

優秀作品 (內容和投影片製作)

• 廖庭煜、陳秉均、汪志剛，儲能整合應用規劃 (pdf，2022 秋作業研究 (上)。清楚說明，問題和數學與程式的對應關係) 
• 王喬誼、潘幸慈、許雅婷：Trading Strategy in the book Python for finance by Yves Hilpisch (pdf，2023 春作業研究 (下)。將一般的方法移到後面。只報告複雜的方法，以便在八分鐘內完成)

成為獨角獸

陳宜伶 編譯，獵頭公司CEO：我面試過3萬多人，這7種人最稀有，商周頭條，2023.11.24

Interesting, Important, and Impactful Operations Management

Gerard P. Cachon, Karan Girotra, Serguei Netessine (2020) Interesting, Important, and Impactful Operations Management. Manufacturing & Service Operations Management 22(1):214-222. 

Machine-Guided Discovery of a Real-World Rogue Wave (瘋狗浪) Model

台北天文館，AI 找到如何預測巨浪的公式，2023 年 12 月 02 日 

Dion Häfner, Johannes Gemmrich, Markus Jochum, Machine-Guided Discovery of a Real-World Rogue Wave Model,  Proceedings of the National Academy of Sciences (2023), 120(48). (arXiv, data, code)

工業系同學的起薪

長期關心青年低薪的問題。根據大規模的數據和個人的經驗，教育是脫離貧窮的有效方法 (1)。

這幾天收到學校的資料，看到這些統計數字，分享一下。

AI 養魚解決老師傅技術失傳問題

邱倢芯，從魚缸到金目鱸養殖場，AI 養魚解決老師傅技術失傳問題，科技新報，2023 年 11 月 09 日

對於建置一套系統，許多業主最擔心的莫過於建置成本；對此，劉建伸坦言，系統本身的確不便宜，從導入初期至今已經投入 8 位數的成本，但後續帶來的效益也相當明顯，像是養殖戶過去都得長時間留守漁塭，在導入系統後，可將平均每天 8 小時的工作時間降低至 6.5 小時。

人力成本也可進一步降低，王靜儀估算，透過導入系統，每 100 公頃養殖面積的人力可從 33 人降低至 10 人，節省近 2,000 萬的人力成本。

(大學剛畢業) 找工作或推甄研究所所需的履歷表和自傳

(一頁的) 履歷表 (註 1)：包括 (如果某項沒有，則不用寫)

• 個人基本資料和聯絡方式：例如電話、email、(專業的) blog 等等。
• 學歷：大學以上的部份。
• 成績：如果前幾名，可以列班級或系排名，學校成績單通常有附帶說明。某同學是全系第一名，卻沒有說明之；找工作時，謙虛不是美德。某同學是全班第二名，卻說前幾名；別人心中會有疑惑，是第 10 名？如果成績不理想，但主修比較高，可以單獨列出；或者說明逐年進步的情況。

Learning an Inventory Control Policy with General Inventory Arrival Dynamics

S Andaz, C Eisenach, D Madeka, K Torkkola, R Jia, D Foster, S Kakade, Learning an Inventory Control Policy with General Inventory Arrival Dynamics, 2023, arXiv preprint arXiv:2310.17168. (Amazon)

In this paper we address the problem of learning and backtesting inventory control policies in the presence of general arrival dynamics -- which we term as a quantity-over-time arrivals model (QOT). We also allow for order quantities to be modified as a post-processing step to meet vendor constraints such as order minimum and batch size constraints -- a common practice in real supply chains. To the best of our knowledge this is the first work to handle either arbitrary arrival dynamics or an arbitrary downstream post-processing of order quantities. Building upon recent work (Madeka et al., 2022) we similarly formulate the periodic review inventory control problem as an exogenous decision process, where most of the state is outside the control of the agent. Madeka et al. (2022) show how to construct a simulator that replays historic data to solve this class of problem. In our case, we incorporate a deep generative model for the arrivals process as part of the history replay. By formulating the problem as an exogenous decision process, we can apply results from Madeka et al. (2022) to obtain a reduction to supervised learning. Finally, we show via simulation studies that this approach yields statistically significant improvements in profitability over production baselines. Using data from an ongoing real-world A/B test, we show that Gen-QOT generalizes well to off-policy data.

Sparse PCA: A New Scalable Estimator Based On Integer Programming

Kayhan Behdin and Rahul Mazumder, Sparse PCA: A New Scalable Estimator Based On Integer Programming, arXiv:2109.11142v2, 2021. (Julia ahd Gurobi code)

We consider the Sparse Principal Component Analysis (SPCA) problem under the well-known spiked covariance model. Recent work has shown that the SPCA problem can be reformulated as a Mixed Integer Program (MIP) and can be solved to global optimality, leading to estimators that are known to enjoy optimal statistical properties. However, current MIP algorithms for SPCA are unable to scale beyond instances with a thousand features or so. In this paper, we propose a new estimator for SPCA which can be formulated as a MIP. Different from earlier work, we make use of the underlying spiked covariance model and properties of the multivariate Gaussian distribution to arrive at our estimator. We establish statistical guarantees for our proposed estimator in terms of estimation error and support recovery. We propose a custom algorithm to solve the MIP which is significantly more scalable than off-the-shelf solvers; and demonstrate that our approach can be much more computationally attractive compared to earlier exact MIP-based approaches for the SPCA problem. Our numerical experiments on synthetic and real datasets show that our algorithms can address problems with up to 20000 features in minutes; and generally result in favorable statistical properties compared to existing popular approaches for SPCA.

Artificial Intelligence and the Future of Universities

Dimitris Bertsimas, Artificial Intelligence and the Future of Universities, Texas A&M Institute of Data Science, 2023.

24小時內從地獄到天堂 黃玉霖的金牌奇蹟滑冰之旅

 李若雯，24小時內從地獄到天堂 黃玉霖的金牌奇蹟滑冰之旅，FB，2023/10/2

中央社

為什麼我們要在意美國？

US Taiwan Watch：美國台灣觀測站 (陳方隅、李可心、李昱孝、許亞傑)，為什麼我們要在意美國？從外交、制度、重大議題全面解析台美關係，聯經，2021

Genius Makers

Cade Metz, Genius Makers: The Mavericks Who Brought AI to Google, Facebook, and the World, Dutton, 2021.

王曉伯譯，AI製造商沒說的祕密: 企業巨頭的搶才大戰如何改寫我們的世界?，時報文化，2022

智慧農業的成功因素

林一平，智慧農業的成功因素，電子時報，2023-08-21

在台灣，農業物聯網感測設備供應商眾多，然而通訊技術和資料傳輸格式卻千差萬別，導致資料在不同系統間的流通和加值應用面臨著困難。

為了解決這一重要問題，農業部於2023年4月27日推出「智慧農業感測資料格式標準與測試規範」。透過推動資料格式的標準化，提高農業物聯網應用領域中資料串接的效率，同時也降低開發成本，推動農業物聯網的深入應用。

機率與統計的應用

生活中到處充滿著不確定性，例如吃飯排隊的等候時間、機台生產的良率、民調統計數字的分析等等。工業系也開設許多相關的課程，例如品質管制、資料分析、智慧製造、實驗設計、機器學習、人工智慧等等，以解決工商業的問題。

許多人修這門課的時候，很痛苦 (1)。除了了解其應用外，建議可以念一些科普的書，增加學習動機：

電路和電子學的產業知識

工業與系統工程學系隸屬於電機資訊學院，所以系上同學們都要修習電路學和電子學相關的課程。在電機系，這些是非常重要的基礎知識；而且，從 IC 設計、晶圓代工、到封裝測試，台灣有許多世界級的領導廠商，所以，值得同學們好好地學習相關的知識。更多資訊。

以下介紹一些科普和整體產業的書籍，希望可以增加同學們的學習動機，並且增加職涯發展的機會：

The Art of Linear Algebra

Kenji Hiranabe, Graphic notes on Gilbert Strang's "Linear Algebra for Everyone"

I tried intuitive visualizations of important concepts introduced in "Linear Algebra for Everyone".

Champion-level drone racing using deep reinforcement learning

Kaufmann, E., Bauersfeld, L., Loquercio, A. et al. Champion-level drone racing using deep reinforcement learning. Nature 620, 982–987 (2023). https://doi.org/10.1038/s41586-023-06419-4

First-person view (FPV) drone racing is a televised sport in which professional competitors pilot high-speed aircraft through a 3D circuit. Each pilot sees the environment from the perspective of their drone by means of video streamed from an onboard camera. Reaching the level of professional pilots with an autonomous drone is challenging because the robot needs to fly at its physical limits while estimating its speed and location in the circuit exclusively from onboard sensors. Here we introduce Swift, an autonomous system that can race physical vehicles at the level of the human world champions. The system combines deep reinforcement learning (RL) in simulation with data collected in the physical world. Swift competed against three human champions, including the world champions of two international leagues, in real-world head-to-head races. Swift won several races against each of the human champions and demonstrated the fastest recorded race time. This work represents a milestone for mobile robotics and machine intelligence, which may inspire the deployment of hybrid learning-based solutions in other physical systems.

Communication Efficient Fair and Robust Federated Learning

Yaodong Yu, Sai Praneeth Karimireddy, Yi Ma, and Michael I. Jordan, Scaff-PD: Communication Efficient Fair and Robust Federated Learning, arXiv:2307.13381.

We present Scaff-PD, a fast and communication-efficient algorithm for distributionally robust federated learning. Our approach improves fairness by optimizing a family of distributionally robust objectives tailored to heterogeneous clients. We leverage the special structure of these objectives, and design an accelerated primal dual (APD) algorithm which uses bias corrected local steps (as in Scaffold) to achieve significant gains in communication efficiency and convergence speed. We evaluate Scaff-PD on several benchmark datasets and demonstrate its effectiveness in improving fairness and robustness while maintaining competitive accuracy. Our results suggest that Scaff-PD is a promising approach for federated learning in resource-constrained and heterogeneous settings.

Optimization of carbon emission reduction paths in the low-carbon power dispatching process

J. Jin, Q. Wen, S. Cheng, Y. Qiu, X. Zhang, and X. Guo, Optimization of carbon emission reduction paths in the low-carbon power dispatching process, Renewable Energy, Volume 188, April 2022, Pages 425-436.

With the development of low-carbon electricity, the scale of wind power is expanding continuously and carbon trading for thermal power is popularized gradually. In this context, the optimal combination of thermal power and wind power needs to be further promoted to build up the synergy of carbon reduction. To solve such low-carbon power dispatching problem in the wind power integrated system imported with carbon trading, this paper firstly presents a distributed robust optimization model. Next, the scenario-based characterization of wind power and allocation methods of initial carbon emission rights are discussed for model solution. Finally, empirical analysis shows that: (1) the proposed model proves to be rational and feasible, which can accomplish a good compromise between economy, environment and robustness of power system, (2) wind power integration dose help carbon reduction ratio to achieve up to 50% with lower operating costs and carbon emissions, while carbon trading is really an effective approach for tapping greater carbon reduction potential of thermal power, and (3) more reasonable proportions of wind power in coping with its inherent uncertainties, and more appropriate cooperation modes of thermal power for dealing with carbon trading unpredictability are determined under the different requirement of carbon reduction.

National energy system optimization modelling for decarbonization pathways analysis

F.A. Plazas-Niño, N.R. Ortiz-Pimiento, E.G. Montes-Páez, National energy system optimization modelling for decarbonization pathways analysis: A systematic literature review, Renewable and Sustainable Energy Reviews, Volume 162, July 2022, 112406.

Energy planning is fundamental to ensure a sustainable, affordable, and reliable energy mix for the future. Energy system optimization models (ESOMs) are the accurate tools to guide decision-making in national energy planning. This article presents a systematic literature review covering the main ESOMs, the input and output data involved, the trends in scenario analysis for decarbonization pathways in national economies, and the challenges associated with energy system optimization modelling. The first part introduces the characterization of ESOMs, showing a trend in modelling focused on long-term, multisector, multiperiod, bottom-up, linear programming, and perfect foresight. Secondly, the analysis shows the intensive data requirements, including future demand profiles, fuel price projections, energy potentials, and techno-economic characteristics of technologies. This review also reveals that decarbonization pathways are the principal objective in energy system optimization modelling, including key drivers such as high-share renewable energy integration, energy efficiency increase, sector coupling, and sustainable transport. The last section presents ten challenges and their corresponding opportunities in research, highlighting the improvement of spatiotemporal resolution, transparency, the inclusion of social aspects, the representation of developing country features, and quality and availability data.

Deep Learning Interviews

Shlomo Kashani and Amir Ivry,  Deep Learning Interviews: Hundreds of fully solved job interview questions from a wide range of key topics in AI, arXiv:2201.00650

use ChatGPT to learn ***

Jessica Shieh, Best practices for prompt engineering with OpenAI API: How to give clear and effective instructions to GPT-3 and Codex, OpenAI. (API)

Esor Huang，ChatGPT提示語進階說明書：通用三層結構＋九個技巧，輕鬆讓AI完成複雜的任務，關鍵評論網，2023/04/28

janus，30年資深美國律師用ChatGPT當他的助手寫訴狀，結果法官、對方律師、甚至連他自己都傻了，techbang，2023年6月10日 (產生不存在的案例)

• 屬於非線性規劃的章節，不知道同學怎麼問的，ChatGPT 給的答案是用決策樹求解？

Carbon dioxide capture, transport and storage supply chains

Viola Becattini, Paolo Gabrielli, Cristina Antonini, Jordi Campos, Alberto Acquilino, Giovanni Sansavini, Marco Mazzotti, Carbon dioxide capture, transport and storage supply chains: Optimal economic and environmental performance of infrastructure rollout, International Journal of Greenhouse Gas Control, Volume 117, June 2022, 103635.

This work presents a novel optimization framework for the optimal design of carbon capture, transport, and storage supply chains in terms of installation, sizing and operation of carbon dioxide (CO2) capture and transport technologies. The optimal design problem is formulated as a mixed-integer linear program that minimizes the total costs of the supply chains while complying with different emissions reduction pathways over a deployment time horizon of 25 years. All design decisions are time-dependent and are taken with a yearly resolution. Whereas the model is general, here its features are illustrated by designing optimal supply chains to decarbonize the Swiss waste-to-energy sector, for various emission reduction pathways, when up to two storage sites are considered, namely one in the North Sea assumed to be already available and a hypothetical one in Switzerland assumed to be possibly available in the future. Findings show that, unless a domestic storage site becomes available soon, the transport cost is the greatest contribution to the overall costs, followed by the capture cost, while the storage cost plays only a minor role. Pipelines are the most cost-effective mode of transport for large volumes of transported CO2, especially when considering multi-year time horizons for the planning of the supply chains. Ship and barge connections are competitive with pipeline connections, whereas rail and truck connections are cost-optimal only when considering shortsighted time horizons or small volumes of CO2 transported.

Queueing Theory: Classical and Modern Methods

Dimitris Bertsimas and David Gamarnik, Queueing Theory: Classical and Modern Methods, ‎Dynamic Ideas, 2022.

STRUCTURE OF THE BOOK:

• Part I describes single and multi-server queues.
• Part II treats single and multiclass queueing networks (MQNETs).
• Part III introduces asymptotic methods, including queueing networks in heavy traffic, large deviations, call centers, queues in space, and the supermarket model.
• Part IV outlines the use of optimization in queueing networks.
• Part V presents Markov chains and processes, Brownian motion, and weak convergence in the Appendix.

Supervised machine learning for theory building and testing

Yen-Chun Chou, Howard Hao-Chun Chuang, Ping Chou, and Rogelio Oliva, Supervised machine learning for theory building and testing: Opportunities in operations management, Journal of Operations Management, 2023. pp. 1–33. (Codes in R)

Machine learning's (ML's) unique power to approximate functions and identify non-obvious regularities in data have attracted considerable attention from researchers in natural and social sciences. The emergence of predictive modeling applications in OM studies notwithstanding, it remains unclear how OM scholars can effectively leverage supervised ML for theory building and theory testing, the primary goals of scientific research. We attempt to fill this gap by conducting a literature review of recent developments in supervised ML in OM to identify vacancies in the extant literature, shedding light on how ML applications can move beyond problem-solving into theory building, and formulating a procedure to help OM scholars leverage ML for exploratory theory development. Our procedure employs the random forest with well-developed properties and inference toolkits that are crucial for empirical research. We then expand the boundary of ML usage and connect supervised ML to the explanatory modeling and hypothesis testing employed by OM empiricists for decades, and discuss the use of supervised ML for causal inference from observational data. We posit that contemporary ML can facilitate pattern exploration and enhance the validity of theory testing. We conclude by discussing directions for future empirical OM studies that aim to leverage ML.

The role of optimization in some recent advances in data-driven decision-making

Baardman, L., Cristian, R., Perakis, G. et al. The role of optimization in some recent advances in data-driven decision-making. Mathematical Programming 200, 1–35 (2023). https://doi.org/10.1007/s10107-022-01874-9.

Data-driven decision-making has garnered growing interest as a result of the increasing availability of data in recent years. With that growth many opportunities and challenges have sprung up in the areas of predictive and prescriptive analytics. Often, optimization can play an important role in tackling these issues. In this paper, we review some recent advances that highlight the difference that optimization can make in data-driven decision-making. We discuss some of our contributions that aim to advance both predictive and prescriptive models. First, we describe how we can optimally estimate clustered models that result in improved predictions. Next, we consider how we can optimize over objective functions that arise from tree ensemble models in order to obtain better prescriptions. Finally, we discuss how we can learn optimal solutions directly from the data allowing for prescriptions without the need for predictions. For all these new methods, we stress the need for good performance but also the scalability to large heterogeneous datasets.

失敗為成功之母？

有時候，很努力也不一定會有成果。

心情低落了好幾天，喝點小酒和親友聊聊天，轉換心情。

友達 AI 數位化打造高效供應鏈

吳珍儀，友達AI數位化打造高效供應鏈 蟬聯美國製造領導獎，Yahoo 財經，2023年6月29日

友達在智慧研發環節開發色彩飽和度模擬系統，引入大數據分析演算法，進行材料選用的相關模擬，提前預測客戶喜好的顏色和材料穿透率，使設計階段效率提升83%。在面板前段製程中，建立元件標準化資料庫並結合佈局設計演算法，成功縮短50%的光罩設計時程。

Automated Machine Learning

Haifeng Jin, François Chollet, Qingquan Song, and Xia Hu. "AutoKeras: An AutoML Library for Deep Learning." the Journal of Machine Learning Research 6 (2023): 1-6. (Download)

To use deep learning, one needs to be familiar with various software tools like TensorFlow or Keras, as well as various model architecture and optimization best practices. Despite recent progress in software usability, deep learning remains a highly specialized occupation. To enable people with limited machine learning and programming experience to adopt deep learning, we developed AutoKeras, an Automated Machine Learning (AutoML) library that automates the process of model selection and hyperparameter tuning. AutoKeras encapsulates the complex process of building and training deep neural networks into a very simple and accessible interface, which enables novice users to solve standard machine learning problems with a few lines of code. Designed with practical applications in mind, AutoKeras is built on top of Keras and TensorFlow, and all AutoKeras-created models can be easily exported and deployed with the help of the TensorFlow ecosystem tooling.

高教的反向重分配現象

RedHerrings，【數字說話】教育能翻轉人生嗎——高教的反向重分配現象，報導者，2018.9.10

一份來自台灣大學經濟學系系主任林明仁與學生沈暉智的研究，揭露台灣高等教育反向重分配的事實：越有錢的人，子女進入台大的機率也越高，而台大花在每位學生的支出明顯也高於其他私立學校，造成有錢人子女得到政府的資源相對多。國立大學長期維持「低學費高補貼」是否真的公平？

畢業系友經驗分享

中原大學工業系

• 梅驊，不要放過每個讓自己成長的機會 (82 級系友，現任街口支付董事長)
• 陳崇正，你想成為什麼樣的人 (2020 中原工工學士，2022 德州農工大學工業與系統工程學系碩士)

我的 YouTube Channel 訂閱數和公益捐款

YouTube 和此部落格的廣告收入，將全數捐出做公益。

• 1630 位訂閱者 (2023/6/10)
• 1260 位訂閱者 (2022/2/12)
• 1000 位訂閱者 (2021/3/18)
• 680 位訂閱者 (2020/3/26)

暑期實習

許多的企業透過軟硬體，提供商品或服務給顧客。工業系強調系統整合，透過資訊流和決策支援系統 (中的演算法)，診斷與改進企業的效能。最近，因為數位轉型和人工智慧的興盛，系上暑期實習中，軟體工作的需求變多。同學們必須在學校中有所準備，企業也才敢將他們的資料開放給你們使用，以開發軟體；配合肯學肯問的精神，才能解決問題、並且學到東西。反之，如果沒有相關的能力、又找不到合適的短期任務，就可能指定一些雜事 (1)。

以下，針對同學們常見的提問，提供一些建議：

Generative AI

S. Huang, P. Grady, and GTP-3, Generative AI: A Creative New World, Sequoia Capital (紅杉資本), September 19, 2022.

A powerful new class of large language models is making it possible for machines to write, code, draw and create with credible and sometimes superhuman results.

The advice I wish I heard at the graduation I never had

Bill Gates, 5 things I wish I heard at the graduation I never had, May 13, 2023

1. The first thing is, your life isn’t a one-act play.
2. The second piece of advice I wish I heard at my graduation is that you are never too smart to be confused.
3. My third piece of advice is to gravitate toward work that solves an important problem.
4. My fourth piece of advice is simple: Don’t underestimate the power of friendship.
5. My last piece of advice is the one I could have used the most. It took me a long time to learn. And it is this: You are not a slacker if you cut yourself some slack. 

Augustin Hadelich

MNA 牛耳藝術

Augustin Hadelich 五歲開始學小提琴，七歲舉辦人生第一場音樂會，然十五歲那年，一場火災意外使他暫停演奏小提琴一整年的時間，臉和握弓的手都嚴重燒傷。家鄉之火、身心苦痛，只是更堅定了他以音樂為一生志業的道路。

再多的獎項，再催淚的故事，都無法明言親聆他琴聲，自心底無法自主竄起的感動和熱情。

Optimization in Online Content Recommendation Services

Omar Besbes, Yonatan Gur, Assaf Zeevi, Optimization in Online Content Recommendation Services: Beyond Click-Through Rates, 18(1), pp. 15–33, Manufacturing & Service Operations Management, Volume 18, Issue 1, Winter 2016. 

A new class of online services allows Internet media sites to direct users from articles they are currently reading to other content they may be interested in. This process creates a “browsing path” along which there is potential for repeated interaction between the user and the provider, giving rise to a dynamic optimization problem. A key metric that often underlies this recommendation process is the click-through rate (CTR) of candidate articles. Whereas CTR is a measure of instantaneous click likelihood, we analyze the performance improvement that one may achieve by some lookahead that accounts for the potential future path of users. To that end, by using some data of user path history at major media sites, we introduce and derive a representation of content along two key dimensions: clickability, the likelihood to click to an article when it is recommended; and engageability, the likelihood to click from an article when it hosts a recommendation. We then propose a class of heuristics that leverage both clickability and engageability, and provide theoretical support for favoring such path-focused heuristics over myopic heuristics that focus only on clickability (no lookahead). We conduct a live pilot experiment that measures the performance of a practical proxy of our proposed class, when integrated into the operating system of a worldwide leading provider of content recommendations, allowing us to estimate the aggregate improvement in clicks per visit relative to the CTR-driven current practice. The documented improvement highlights the importance and the practicality of efficiently incorporating the future path of users in real time.

2019 澤倫斯基當選後，親俄作為不受俄方理睬

王宏恩，2019澤倫斯基當選也被視為是緩和，然後俄羅斯就進攻了，思想坦克，2023 年 4 月 27 日 

時間到了2019年的總統選舉，Poroshenko的政府內爆發數起重大弊案，聲望持續下滑。此時一個全新的政黨崛起：人民公僕黨（Sluha Narodu, Servant of the People）。這個黨名當初就是抄襲著名電視劇，2018年才成立，也提名了喜劇演員澤倫斯基出馬角逐總統選舉。澤倫斯基完全沒有從政的經驗，也沒有提出任何鮮明的政策，但他有一個非常鮮明的口號──「打倒貪腐，不要戰爭」。

澤倫斯基在各地競選時，都認為烏克蘭當前最主要的問題就是政府的貪汙，而不是其他問題，而只有他這個素人當選才有辦法打敗貪汙、對抗深層政府。另一方面，由於烏克蘭東部頓巴斯戰爭已經持續數年，消耗烏克蘭國力巨大，因此澤倫斯基也堅持說要以對話代替武力衝突，他不理解為何當前執政黨不跟對方對話，認為只要好好對話就能夠創造和平。...

壓力管理

學校的導師會議，邀請郭晏汝諮商心理師來演講「學生自傷問題」。推薦其壓力管理的通識課，課程教學目標

1. 認識產生壓力的外在因子。
2. 瞭解、辨識與覺察壓力對個人認知、情緒、生理的優劣性影響。
3. 從各項活動體驗感知與體現個人因應壓力的人際/行為模式。
4. 從全人發展角度學習有效的壓力管理原則與技巧。

A Practical End-to-End Inventory Management Model with Deep Learning

Meng Qi, Yuanyuan Shi, Yongzhi Qi, Chenxin Ma, Rong Yuan, Di Wu, Zuo-Jun (Max) Shen (2023) A Practical End-to-End Inventory Management Model with Deep Learning. Management Science 69(2):759-773. (Data and Python codes) 

We investigate a data-driven multiperiod inventory replenishment problem with uncertain demand and vendor lead time (VLT) with accessibility to a large quantity of historical data. Different from the traditional two-step predict-then-optimize (PTO) solution framework, we propose a one-step end-to-end (E2E) framework that uses deep learning models to output the suggested replenishment amount directly from input features without any intermediate step. The E2E model is trained to capture the behavior of the optimal dynamic programming solution under historical observations without any prior assumptions on the distributions of the demand and the VLT. By conducting a series of thorough numerical experiments using real data from one of the leading e-commerce companies, we demonstrate the advantages of the proposed E2E model over conventional PTO frameworks. We also conduct a field experiment with JD.com, and the results show that our new algorithm reduces holding cost, stockout cost, total inventory cost, and turnover rate substantially compared with JD’s current practice. For the supply chain management industry, our E2E model shortens the decision process and provides an automatic inventory management solution with the possibility to generalize and scale. The concept of E2E, which uses the input information directly for the ultimate goal, can also be useful in practice for other supply chain management circumstances.

Model-Based Deep Learning

N. Schlezinger, Y. Eldar, and S. Boyd, Model-Based Deep Learning: On the Intersection of Deep Learning and Optimization, IEEE Access, vol. 10, 2022.

Decision making algorithms are used in a multitude of different applications. Conventional approaches for designing decision algorithms employ principled and simplified modelling, based on which one can determine decisions via tractable optimization. More recently, deep learning approaches that use highly parametric architectures tuned from data without relying on mathematical models, are becoming increasingly popular. Model-based optimization and data-centric deep learning are often considered to be distinct disciplines. Here, we characterize them as edges of a continuous spectrum varying in specificity and parameterization, and provide a tutorial-style presentation to the methodologies lying in the middle ground of this spectrum, referred to as model-based deep learning. We accompany our presentation with running examples in super-resolution and stochastic control, and show how they are expressed using the provided characterization and specialized in each of the detailed methodologies. The gains of combining model-based optimization and deep learning are demonstrated using experimental results in various applications, ranging from biomedical imaging to digital communications.

Using AI to Accelerate Scientific Discovery

Demis Hassabis, Using AI to Accelerate Scientific Discovery, Institute for Ethics in AI Oxford, 2022.

A Survey of Quantization Methods for Efficient Neural Network Inference

Amir Gholami, Sehoon Kim, Zhen Dong, Zhewei Yao, Michael W. Mahoney, Kurt Keutzer, A Survey of Quantization Methods for Efficient Neural Network Inference, arXiv:2103.13630v3. 

As soon as abstract mathematical computations were adapted to computation on digital computers, the problem of efficient representation, manipulation, and communication of the numerical values in those computations arose. Strongly related to the problem of numerical representation is the problem of quantization: in what manner should a set of continuous real-valued numbers be distributed over a fixed discrete set of numbers to minimize the number of bits required and also to maximize the accuracy of the attendant computations? This perennial problem of quantization is particularly relevant whenever memory and/or computational resources are severely restricted, and it has come to the forefront in recent years due to the remarkable performance of Neural Network models in computer vision, natural language processing, and related areas. Moving from floating-point representations to low-precision fixed integer values represented in four bits or less holds the potential to reduce the memory footprint and latency by a factor of 16x; and, in fact, reductions of 4x to 8x are often realized in practice in these applications. Thus, it is not surprising that quantization has emerged recently as an important and very active sub-area of research in the efficient implementation of computations associated with Neural Networks. In this article, we survey approaches to the problem of quantizing the numerical values in deep Neural Network computations, covering the advantages/disadvantages of current methods. With this survey and its organization, we hope to have presented a useful snapshot of the current research in quantization for Neural Networks and to have given an intelligent organization to ease the evaluation of future research in this area.

Geoff Hinton on revolutionizing artificial intelligence ... again

The Robot Brains Podcast with Pieter Abbeel, Geoff Hinton on revolutionizing artificial intelligence ... again

Corporate spending on AI and 2 nice reports

Inside AI  (Could subscribe to free newsletters)

學習動力與方向 (2/2)

如同如何準備研究所中所言，不一定要念研究所。在網路時代，知識的取得已經非常方便，例如 edX 或 Coursera，所以大學部的基礎知識和自學的動力很重要 (1)。以 The Analytics Edge 而言，是 MIT 商業分析 (Business Analytics) 碩士的必修課，該碩士學費一年七萬美金，所以此線上免費課程價值一萬美金。

Software Engineering at Google

Titus Winters, Tom Manshreck, and Hyrum Wright, Software Engineering at Google, O'Reilly Media, March 2020. (Read online)

The Software Engineering at Google book (“SWE Book”) is not about programming, per se, but about the engineering practices utilized at Google to make their codebase sustainable and healthy. (These practices are paramount for common infrastructural code such as Abseil.)

The Effective Executive

Peter F. Drucker, The Effective Executive: The Definitive Guide to Getting the Right Things Done, HarperCollins Publishers, January 3, 2006.

齊若蘭譯，杜拉克談高效能的5個習慣，遠流，2009

The measure of the executive, Peter F. Drucker reminds us, is the ability to "get the right things done." This usually involves doing what other people have overlooked as well as avoiding what is unproductive. Intelligence, imagination, and knowledge may all be wasted in an executive job without the acquired habits of mind that mold them into results.

Yoshua Bengio joins Host Pieter Abbeel

The Robot Brains Podcast, S3 E1 Yoshua Bengio joins Host Pieter Abbeel: LLMs, Cognition, Causality, Responsible AI, Creativity

01:17:50 - brain is more creative when walking

2022 Franz Edelman Award

 2022 Edelman Competition (videos)

Leonardo J. Basso et al., Analytics Saves Lives During the COVID-19 Crisis in Chile, INFORMS Journal on Applied Analytics, 2023, 53(1):9-31. (2022 Franz Edelman Award) (statistical analysis, integer programming, regression)

During the COVID-19 crisis, the Chilean Ministry of Health and the Ministry of Sciences, Technology, Knowledge and Innovation partnered with the Instituto Sistemas Complejos de Ingeniería (ISCI) and the telecommunications company ENTEL, to develop innovative methodologies and tools that placed operations research (OR) and analytics at the forefront of the battle against the pandemic. These innovations have been used in key decision aspects that helped shape a comprehensive strategy against the virus, including tools that (1) provided data on the actual effects of lockdowns in different municipalities and over time; (2) helped allocate limited intensive care unit (ICU) capacity; (3) significantly increased the testing capacity and provided on-the-ground strategies for active screening of asymptomatic cases; and (4) implemented a nationwide serology surveillance program that significantly influenced Chile’s decisions regarding vaccine booster doses and that also provided information of global relevance. Significant challenges during the execution of the project included the coordination of large teams of engineers, data scientists, and healthcare professionals in the field; the effective communication of information to the population; and the handling and use of sensitive data. The initiatives generated significant press coverage and, by providing scientific evidence supporting the decision making behind the Chilean strategy to address the pandemic, they helped provide transparency and objectivity to decision makers and the general population. According to highly conservative estimates, the number of lives saved by all the initiatives combined is close to 3,000, equivalent to more than 5% of the total death toll in Chile associated with the pandemic until January 2022. The saved resources associated with testing, ICU beds, and working days amount to more than 300 million USD.

AI 溝通師

邱韞蓁編譯，不寫程式，也能拿千萬年薪！AI大浪催生新職業「AI溝通師」，商周頭條，2023.03.03 

1. AI 大戰鳴槍，催生熱門職業：「AI 溝通師 (prompt engineer)」。由 OpenAI 前員工創立的新創公司 Anthropic，不要求應徵者有電腦科學學位，還開出上看33萬美元 (折合新台幣約1010萬) 的高薪。
2. AI 溝通師不僅要具備機器學習、自然語言處理等技術面硬實力，創造力、清楚表達、合作等軟實力也不可少。他們負責創造或改善輸入給 AI 的指令 (prompt)，讓結果更好。
3. 除了科技業，連醫療、法律、遊戲業都開始徵才！對於一般人而言，也能到 PromptBase、Fiverr 等線上平台，花2美元購買 AI 指令集。

Exploring the Whole Rashomon Set of Sparse Decision Trees

Rui Xin, Chudi Zhong, Zhi Chen, Takuya Takagi, Margo Seltzer, Cynthia Rudin, Exploring the Whole Rashomon Set of Sparse Decision Trees, NeurIPS (oral), 2022. (code) | (bib) | (5 min video)

In any given machine learning problem, there may be many models that could explain the data almost equally well. However, most learning algorithms return only one of these models, leaving practitioners with no practical way to explore alternative models that might have desirable properties beyond what could be expressed within a loss function. The Rashomon set is the set of these all almost-optimal models. Rashomon sets can be extremely complicated, particularly for highly nonlinear function classes that allow complex interaction terms, such as decision trees. We provide the first technique for completely enumerating the Rashomon set for sparse decision trees; in fact, our work provides the first complete enumeration of any Rashomon set for a non-trivial problem with a highly nonlinear discrete function class. This allows the user an unprecedented level of control over model choice among all models that are approximately equally good. We represent the Rashomon set in a specialized data structure that supports efficient querying and sampling. We show three applications of the Rashomon set: 1) it can be used to study variable importance for the set of almost-optimal trees (as opposed to a single tree), 2) the Rashomon set for accuracy enables enumeration of the Rashomon sets for balanced accuracy and F1-score, and 3) the Rashomon set for a full dataset can be used to produce Rashomon sets constructed with only subsets of the data set. Thus, we are able to examine Rashomon sets across problems with a new lens, enabling users to choose models rather than be at the mercy of an algorithm that produces only a single model.

關於我和鬼變成家人的那件事

恐同男警吳明翰 (許光漢 飾) ，誤撿地上紅包，沒想到紅包裡的對象是個男的 (林柏宏 飾) ！被迫男男冥婚的明翰，一路衰到底，不但甩不掉冥婚對象，就連警花林子晴 (王淨 飾) 埋線已久的緝毒案，都被他搞砸。為了挽救危機，恐同又怕鬼的明翰，別無選擇，即使人鬼殊途也要和鬼老公毛毛攜手跨界追兇，一場荒謬絕倫、笑中帶淚的旅程就此展開！

台灣經濟史四百年

 吳聰敏，台灣經濟史四百年，春山出版，2023。

本書講述台灣經濟400年的故事，我所選擇的議題主要是以往我曾經做過研究的。毫無疑問，400年來還有很多重要的議題，但限於我個人的能力有限，本書並沒有講到。例如，晚清到日治初期的高山原住民，晚清的茶業發展，閩南人與客家人在經濟發展中的角色，日治時期在台日本人的發展，戰後的土地改革，黨國資本主義等。我期待未來有人能從經濟的角度，把這些故事寫出來。

大三機器學習的教學評量

身分證檢查程式

無論是程式設計課程或是畢業專題，我都會用這一個程式 (*)，請同學們實作；可以貫穿控制結構和多種資料結構。順便說明，我的朋友如何利用檢查和 (checksum) 的概念，賣一套一百多萬的加油換點數資訊系統，希望激起同學們的學習動機。

打造高效AI推薦系統

 蔡銘仁，打造高效AI推薦系統　林永隆率創鑫智慧挺進世界，EE Times Taiwan，2022-10-27

新創公司創鑫智慧僅成軍第三年，首款人人工智慧(AI)加速晶片就採用成本高昂的台積電7nm製程，吸引業界關注；董事長暨執行長林永隆在半導體業界累積近40年的專業資歷，更讓外界對公司的前景抱有高度期待。他們擘劃的宏大願景，是立志成為世界級的AI加速器供應商。...

思維的製程

彭建文，思維的製程：台積電教我的思維進階法，練成全局經營腦和先進工作術，商業周刊，2023

職場不怕碰上難題，怕的是不會聰明解決。

學習台積電多年淬鍊的系統性問題解決策略，

你也能優化思維的製程，扎穩職涯腳步，累積世界第一競爭力。

What makes us happy in life?

Marc Schulz and Robert Waldinger, An 85-year Harvard study found the No. 1 thing that makes us happy in life: It helps us ‘live longer’,  CNBC, Feb 10, 2023.

Contrary to what you might think, it’s not career achievement, money, exercise, or a healthy diet. The most consistent finding we’ve learned through 85 years of study is: Positive relationships keep us happier, healthier, and help us live longer. Period. 

Multimodal artificial intelligence

Jessica Leung, Omega Rho Keynote: Artificial Intelligence and the Future of Universities, ORMS Today, 2022.

Léonard Boussioux, Cynthia Zeng, Théo Guénais, and Dimitris Bertsimas, Hurricane Forecasting: A Novel Multimodal Machine Learning Framework, Weather and Forecasting, March 2022, 37(6), pp. 817–831.

Soenksen, L.R., Ma, Y., Zeng, C. et al. Integrated multimodal artificial intelligence framework for healthcare applications. Nature Machine Intelligence 5, 149 (2022). https://doi.org/10.1038/s41746-022-00689-4. (Data and Code)

學習大數據 (big data) 的技能

一些工具或念個學位

可以參考 DS Examiner, Data Scientist Foundations: The Hard and Human Skills You Need, November 8, 2013

或者  Insight Data Science Fellows Program 說明了可能使用的工具

1. Software Engineering Best Practices: Learn how to contribute to a large code-base and instrument a web application to collect data. Tools you may learn: Python, Git, LAMP web stack, Javascript, Flask.
2. Storing and Retrieving Data: How to clean data, store it in the appropriate database or distributed data storage system and then run queries to retrieve the information needed for analysis. Tools you may will learn: MySQL, Hadoop, Hive.
3. Statistical Analysis & Machine Learning: Learn industry best practices for doing basic and advanced statistical analysis on large data sets. Tools you may learn: R, NumPy & SciPy, Mahout.
4. Visualizing and Communicating Results: Learn how to effectively communicate your findings visually and verbally. Tools you may learn: D3 Javascript library, visualization and presentation best practices. 

Generalized Synthetic Control for TestOps at ABI

Luis Costa, Vivek F. Farias, Patricio Foncea, Jingyuan (Donna) Gan, Ayush Garg, Ivo Rosa Montenegro, Kumarjit Pathak, Tianyi Peng, and Dusan Popovic, Generalized Synthetic Control for TestOps at ABI: Models, Algorithms, and Infrastructure, To appear in INFORMS Journal on Applied Analytics (Winner, Daniel H. Wagner Prize 2022)

We describe a novel optimization-based approach– Generalized Synthetic Control (GSC)– to learning from experiments conducted in the world of physical retail. GSC solves a long-standing problem of learning from physical retail experiments when treatment effects are small, the environment is highly noisy and non-stationary, and interference and adherence problems are commonplace. The use of GSC has been shown to yield an approximately 100x increase in power relative to typical inferential methods and forms the basis of a new large-scale testing platform: ‘TestOps’. TestOps was developed and has been broadly implemented as part of a collaboration between Anheuser Busch Inbev (ABI) and an MIT team of operations researchers and data engineers. TestOps currently runs physical experiments impacting approximately 135M USD in revenue every month and routinely identifies innovations that result in a 1-2% increase in sales volume. The vast majority of these innovations would have remained unidentified absent our novel approach to inference: prior to our implementation, statistically significant conclusions could be drawn on only ∼ 6% of all experiments; a fraction that has now increased by over an order of magnitude.

Bridging physics-based and data-driven modeling for COVID-19 forecasting

Rui Wang, Danielle Robinson, Christos Faloutsos, Yuyang Wang, and Rose Yu, AutoODE: Bridging physics-based and data-driven modeling for COVID-19 forecasting, NeurIPS 2020 Workshop on Machine Learning in Public Health. (best paper award at the NeurIPS Machine Learning in Public Health Workshop)

As COVID-19 continues to spread, accurately forecasting the number of newly infected, removed and death cases has become a crucial task in public health. While mechanics compartment models are widely-used in epidemic modeling, data-driven models are emerging for disease forecasting. In this work, we investigate these two types of methods for COVID-19 forecasting. Through a comprehensive study, we find that data-driven models outperform physics-based models on the number of death cases prediction. Meanwhile, physics-based models have superior performances in predicting the number of infected and removed cases. In addition, we present an hybrid approach, AutoODE, that obtains a 57.4% reduction in mean absolute errors of the 7-day ahead COVID-19 trajectories prediction compared with the best deep learning competitor.

2021 Franz Edelman Award

2021 Edelman Competition

Koen Peters, Sérgio Silva, Tim Sergio Wolter, Luis Anjos, Nina van Ettekoven, Éric Combette, Anna Melchiori, Hein Fleuren, Dick den Hertog, Özlem Ergun (2022) UN World Food Programme: Toward Zero Hunger with Analytics. INFORMS Journal on Applied Analytics 52(1):8-26. https://doi.org/10.1287/inte.2021.1097 (2021 Franz Edelman Award, WFP received the Nobel Peace Prize in 2020.)

10 Breakthrough Technologies 2023

 David Rotman, 10 Breakthrough Technologies 2023, MIT Technology Review, January 9, 2023.

Our annual look at 10 Breakthrough Technologies—including CRISPR for high cholesterol, battery recycling, AI that makes images, and the James Webb Space Telescope—that will have a profound effect on our lives. Plus care robots, 3-D printing pioneers, and chasing bugs on the blockchain.

Feature selection repository scikit-feature in Python

J. Li, K. Cheng, S. Wang, F. Morstatter, R.P. Trevino, J. Tang, and H. Liu, “Feature Selection: A Data Perspective,” ACM Computing Surveys, vol. 50, no. 6, pp. 1–45, December 2017. https://doi.org/10.1145/3136625 (Python at GitHub)

scikit-feature is an open-source feature selection repository in Python developed by Data Mining and Machine Learning Lab at Arizona State University. It is built upon one widely used machine learning package scikit-learn and two scientific computing packages Numpy and Scipy. scikit-feature contains around 40 popular feature selection algorithms, including traditional feature selection algorithms and some structural and streaming feature selection algorithms.

It serves as a platform for facilitating feature selection application, research and comparative study. It is designed to share widely used feature selection algorithms developed in the feature selection research, and offer convenience for researchers and practitioners to perform empirical evaluation in developing new feature selection algorithms.