, , , ,

SRG paper wins Best Paper Award at ALA 2024 workshop

The paper titled "A Meta-Game Evaluation Framework for Deep Multiagent Reinforcement Learning," authored by Zun Li, while he was a PhD student with SRG, and Michael Wellman just won the best paper award at the 16th Adaptive and Learning Agents…

A Meta-Game Evaluation Framework for Deep Multiagent Reinforcement Learning

Z Li and MP Wellman 33rd International Joint Conference on Artificial Intelligence (IJCAI), August 2024. Forthcoming. Best Paper Award at 16th Adaptive and Learning Agents (ALA) Workshop at AAMAS, May 2024. Abstract Evaluating deep multiagent…

Co-Learning Empirical Games and World Models

MO Smith and MP Wellman 1st Reinforcement Learning Conference (RLC), August 2024. Forthcoming. Abstract Game-based decision-making involves reasoning over both world dynamics and strategic interactions among the agents. Typically, empirical…

Fraud Risk Mitigation in Real-Time Payments: A Strategic Agent-Based Analysis

K Mayo, N Grabill, and MP Wellman 33rd International Joint Conference on Artificial Intelligence (IJCAI), August 2024. Forthcoming. Abstract Whereas standard financial mechanisms for payment may take days to finalize, real-time payments (RTPs)…

Numerix “Trading Tomorrow” Podcast

Michael Wellman was interviewed by James Jockle from Numerix about the implications of AI on financial markets, including market manipulation and regulation. https://www.buzzsprout.com/2243477/14722606

Generalized Response Objectives for Strategy Exploration in Empirical Game-Theoretic Analysis

Y Wang and MP Wellman 23nd International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS), May 2024. Forthcoming. Abstract In the policy-space response oracle (PSRO) framework, strategy sets defining an empirical game are iteratively…
, , , ,

Zun Li defends dissertation

  On 17 January 2024, Zun Li successfully defended his PhD dissertation titled, "Artificial Intelligence Algorithms for Large Economic and Computer Games." Thanks to the dissertation committee members: Michael Wellman (chair) …

Learning to Manipulate a Financial Benchmark

M Shearer, G Rauterberg, and MP Wellman Proceedings of 4th ACM International Conference on AI in Finance (ICAIF'23), pages 592–600, November 2023. Abstract Financial benchmarks estimate market values or reference rates used in a wide variety…

Strategic Knowledge Transfer

MO Smith, T Anthony, MP Wellman Journal of Machine Learning Research (JMLR), 24(233):1−96, August 2023. Abstract In the course of playing or solving a game, it is common to face a series of changing other-agent strategies. These strategies…

Learning to play against any mixture of opponents

MO Smith, T Anthony, MP Wellman Frontiers of Artificial Intelligence (Sec. Machine Learning and Artificial Intelligence), Volume 6, July 2023. Abstract Intuitively, experience playing against one mixture of opponents in a given domain should…
, , , , , ,

Katherine Mayo defends dissertation proposal

On July 17, 2022, Katherine Mayo presented and successfully defended her dissertation proposal titled "A Strategic Analysis of Economic and Technological Changes in Financial Networks Using Agent-Based Modeling." The dissertation committee…

Learning Parameterized Families of Games

M Gatchel and B Wiedenbeck 22nd International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS), pp. 1044–1052, June 2023. Abstract To understand the impact of parameters in strategic environments, typical game-theoretic analysis…

Empirical Game-Theoretic Analysis for Mean Field Games

Y Wang and MP Wellman 22nd International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS), pp. 1025–1033, June 2023. Abstract We present a simulation-based approach for solution of mean field games (MFGs), using the framework…
, , , , , ,

Christine Konicki defends dissertation proposal

On October 17, 2022, Christine Konicki presented and successfully defended her dissertation proposal titled "A Tree-Exploiting Approach to Empirical Game-Theoretic Analysis for Extensive-Form Games." The dissertation committee comprises: …
, , , ,

Machine Learning, Algorithmic Trading, and Manipulation

Columbia Blue Sky Blog post based on the report by Megan Shearer, Gabriel Rauterberg, and Michael Wellman.
, , , , , ,

Megan Shearer defends dissertation

On June 27, 2022, Megan Shearer successfully defended her PhD dissertation titled, "Modeling Trading Strategies in Financial Markets with Data, Simulation, and Deep Reinforcement Learning." Congratulations, Dr. Shearer! We wish you a bright…

Solving Structured Hierarchical Games Using Differential Backward Induction

Z Li, F Jia, A Mate, S Jabbari, M Chakraborty, M Tambe, and Y Vorobeychik 38th Conference on Uncertainty in Artificial Intelligence (UAI), PMLR 180: pp. 1107–1117, August 2022 Previous version presented at ICLR Workshop on Gamification…
, , , , ,

Yongzhao Wang defends dissertation proposal

On 10 May 2022, Yongzhao Wang presented and successfully defended his dissertation proposal titled "Multi-agent Learning by Iterative Refinement of Game Models." The dissertation committee comprises: Michael Wellman [chair] Tilman…
, , , ,

Zun Li defends dissertation proposal

On 27 April 2022, Zun Li presented and successfully defended his dissertation proposal titled "A Modern AI Approach to Strategic Reasoning over Complex Multiagent Systems". The dissertation committee comprises: Michael Wellman [chair] …

Evaluating Strategy Exploration in Empirical Game-Theoretic Analysis

Y Wang, Q Ma and MP Wellman 21st International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS), pages 1346—1354, May 2022. Abstract In empirical game-theoretic analysis (EGTA), game models are extended iteratively through…

Stability Effects of Arbitrage in Exchange Traded Funds: An Agent-Based Model

M Shearer, D Byrd, TH Balch, and MP Wellman 2nd ACM International Conference on AI in Finance (ICAIF), Article No.: 49, pages 1–9, November 2021. Abstract An index-based exchange traded fund (ETF) with underlying securities that trade on…

Timing is Money: The Impact of Arrival Order in Beta-Bernoulli Prediction Markets

B Martin, S Kutty, and M Chakraborty 2nd ACM International Conference on AI in Finance (ICAIF), Article No.: 41, pages 1–9, November 2021. Abstract Prediction markets are incentive-based mechanisms for eliciting and combining the diffused,…

An Agent-Based Model of Strategic Adoption of Real-Time Payments

K Mayo, S Fozdar, and MP Wellman 2nd ACM International Conference on AI in Finance (ICAIF), Article No.: 45, pages 1–9, November 2021. Abstract Real-time payments (RTPs) allow consumers to receive funds before the completion of payment…

Building Action Sets in a Deep Reinforcement Learner

Y Wang, A Sinha, S CH-Wang, and MP Wellman 20th IEEE International Conference on Machine Learning and Applications (ICMLA-21), pages 484–489, December 2021. Abstract In many policy-learning applications, the agent may execute a set of actions…

A Strategic Analysis of Portfolio Compression

K Mayo and MP Wellman 2nd ACM International Conference on AI in Finance (ICAIF), Article No.: 20, pages 1–8, November 2021. Extended abstract appeared in 20th International Conference on Autonomous Agents and Multiagent Systems (AAMAS),…

An Agent-Based Model of Financial Benchmark Manipulation

M Shearer, G Rauterberg, and MP Wellman ICML Workshop on Applications and Infrastructure for Multi-Agent Learning, June 2019 Abstract Financial benchmarks estimate market values or reference rates used in a wide variety of contexts, but are…

Learning-Based Trading Strategies in the Face of Market Manipulation

X Wang, C Hoang, and MP Wellman ACM International Conference on AI and Finance, October 2020. Abstract We study learning-based trading strategies in markets where prices can be manipulated through spoofing: the practice of submitting spurious…

Economic reasoning from simulation-based game models

MP Wellman Œconomia, 10(2):257–278, 2020. Abstract Simulation modeling in economics has historically been viewed as an alternative to mainstream analytic technique, and as such has generally and intentionally avoided the focus on rational…

Market manipulation: An adversarial learning framework for detection and evasion

X Wang and MP Wellman 29th International Joint Conference on Artificial Intelligence, Special Track on AI in FinTech, pages 4626–4632, 2020. Abstract We propose an adversarial learning framework to capture the evolving game between a regulator…
, , ,

interview with Grainstone Lee

I spoke with Simon Grainger recently for his firm's "Views from Academia" series. The interview is based on presentations I've made recently on "Game Playing meets Game Theory", discussing the connection between game-theoretic reasoning and…

Cap-and-trade emissions regulation: A strategic analysis

F Cheng , Y Engel, and MP Wellman Proceedings of the 28th International Joint Conference on Artificial Intelligence, pages 187–193, August 2019. Abstract Cap-and-trade schemes are designed to achieve target levels of regulated emissions…
, , ,

My interview with Bill Powers on AI Decision Makers

https://youtu.be/SnTf-iWUTpk Recorded June 2018, as part of a series on Machine Behavior, in conjunction with publication of a position paper in Nature on the topic.

Multiscale Network Games of Collusion and Competition

Principal Investigators Michael Wellman (U Michigan) Mingyan Liu (U Michigan) PIs at partner universities: Milind Tambe (University of Southern California) David Kempe (University of Southern California) P. Jeffrey Brantingham…

A Cloaking Mechanism to Mitigate Market Manipulation

X Wang, Y Vorobeychik, and MP Wellman 27th International Joint Conference on Artificial Intelligence, pages 541–547, July 2018. Abstract We propose a cloaking mechanism to deter spoofing, a form of manipulation in financial markets. The…

Incentivizing rider time-shift in a multi-leg public transportation system

M Shearer and MP Wellman 10th International Workshop on Agents in Traffic and Transportation, July 2018. Abstract We develop an incentive scheme for a hub-to-shuttle campus transit system, encouraging riders to shift travel times to improve…
, , ,

Megan Shearer Passes Prelim

Megan Shearer passed her prelim exam, based on her directed study project: Incentivizing Rider Time-Shift in a Multi-Leg Public Transportation System. Congratulations, Megan.
, ,

Asimov Debate on Artificial Intelligence

I had the privilege of participating in the 2018 Isaac Asimov Memorial Debate, on the subject of Artificial Intelligence, held at the American Museum of Natural History on 13 Feb. The event was hosted by Neil deGrasse Tyson, and featured five…

Detecting Financial Market Manipulation: An Integrated Data- and Model-Driven Approach

Principal Investigators Michael Wellman (U Michigan) Uday Rajan (U Michigan) Michael Barr (U Michigan) PIs at partner universities: Tucker Balch (Georgia Tech) Sponsored by the NSF BIGDATA program, grant IIS-1741190. Project…
, , ,

New project on market manipulation

Collaboration with Tucker Balch (Ga Tech), Uday Rajan (UMich), and Michael Barr (UMich), funded by NSF BIGDATA program. CSE news posting.

Accounting for strategic response in an agent-based model of financial regulation

F Cheng and MP Wellman Proceedings of the 18th ACM Conference on Economics and Computation, pages 187–203, June 2017. Abstract Due to complex interactions in financial markets, financial regulations can sometimes produce unexpected outcomes,…
, , , ,

Algocracy Podcast

Uday Rajan and I appeared on the Algocracy and Transhumanism Project Podcast, hosted by John Danaher, to talk about our recent article on the ethics of automated trading.
, ,

Bloomberg article about ASU Origins workshop on adverse AI outcomes

Last weekend I attended a very interesting meeting at Arizona State U, hosted by the Origins Project, devoted to discussions on potential adverse outcomes from AI, and how to avoid them. Bloomberg News ran an article describing it at high level.…

Ethical issues for autonomous trading agents

MP Wellman and U Rajan Minds and Machines, 27:609-624, 2017. Abstract The rapid advancement of algorithmic trading has demonstrated the success of AI automation, as well as gaps in our understanding of the implications of this technology…
Erik Brinkman
, , , ,

Erik Brinkman Defends Thesis Proposal

Erik Brinkman successfully defended his thesis proposal today: Understanding Financial Market Behavior through Empirical Game-Theoretic Analysis. Congratulations, Erik.

JAAMAS Special issue on Autonomous Agents for ABM

Just published: the special issue of J. Autonomous Agents and Multiagent Systems, on Autonomous Agents for Agent-Based Modeling.  Co-editors: Virginia Dignum, Nigel Gilbert, Michael Wellman.

AI Scientist Letter on Autonomous Weapons

Signing the AI Researcher Open Letter on Autonomous Weapons was something of a no-brainer for me.  I am not confident that a ban would be easy to define or enact or enforce, but the attempt is worthwhile if only to give some pause and shine…

Economic Reasoning and AI

A review article on Econ/AI co-authored by David Parkes and myself just came out in the 17 July issue of Science magazine. See also David’s interview about it on phys.org.

CFP: JAAMAS special issue on ABM

Autonomous Agents for Agent-Based Modeling A special issue of the Journal of Autonomous Agents and Multiagent Systems

Strategic Modeling of Dynamic Credit Networks

Principal Investigator Michael Wellman Students Frank Cheng Junming Liu (MS ECE, 2015) Project Goals The 2008 financial crisis demonstrated that complex and opaque networks of credit relationships among firms can set the…

Putting the Agent in Agent-Based Modeling

Article version of my AAMAS-14 keynote talk is now available.

Another cnbc.com Rebuttal

Rishi Narang has written another article on cnbc.com defending charges of HFT front-running, this one a direct attack on me.  Titled “Exposing the falsehood of a prominent HFT critic’s arguments“, Narang attempts to refute my rebuttal…

HFT and Front Running

CNBC recently ran a commentary by Rishi K. Narang, under the headline “High-frequency traders can’t front-run anyone“, in which the author calls our characterizations of HFT “blatantly false”. My rebuttal here.

Arguments about “front running”

One of the brilliant rhetorical devices deployed by Michael Lewis in his public interviews about Flash Boys is referring to some HFT practices as "legal front running". By inserting the "legal", he takes off the table any accusations of lawbreaking.…

“The Myths around Latency Arbitrage”

It was predictable that wading into the public debates around high-frequency trading would attract some attacks on our research.  Nevertheless, it is still novel for me to read a blog claiming that our “research makes a number of basic mistakes”,…

Does US HFT need stricter regulatory oversight?

YES. My “head-to-head” opinion piece in International Financial Law Review. local PDF copy

CNN Money article on latency arbitrage

Quotes our favorite remedy… http://finance.fortune.cnn.com/2013/08/30/latency-arbitrage-costs/


“High frequency trading” in German.  Die Zeit article quoting me in translation.

Every day, another flash crash

column by Michael Brush, MSN Money Bringing attention to “mini flash crashes” reportedly occurring with some regularity in recent months. Quotes me about vulnerability of the markets near beginning of the column.

Michigan Sells 2-Second Advance Version of Consumer Confidence Survey

The University of Michigan and Thomson Reuters are getting a lot of heat from recent reports about a million-dollar deal whereby Michigan releases a data feed with survey results a full two seconds before the results are available to select…

Michigan AI Lab Mini-Symposium Scheduled for 2 May

A showcase of research from across the AI Lab. Start of an annual tradition? See the details.

Credit networks paper presented at Allerton conference

An Empirical Game-Theoretic Analysis of Credit Network Formation
, ,

Quang Duong Defends Dissertation

Quang Duong successfully defended his thesis on 20 July.  Quang's dissertation, entitled Graphical Multiagent Models, demonstrates how to exploit the flexibility and power of probabilistic graphical models for representing and reasoning about…

2012 Trading Agent Competition Results

TAC-12 is now (mostly) history. The Ad Auction (AA) and Supply Chain Management (SCM) divisions had their semifinals and finals in Valencia earlier this month. Results are posted on the TAC web site. The PowerTAC finals were postponed to September. Our…

The Microsecond Market

IEEE Spectrum has a nice feature article (written by David Schneider) in their June 2012 issue about high-frequency trading and the latency arms race in particular.  The article quotes me advocating discrete-time markets.

Going to Liverpool

Visiting the CS dept, will talk about EGTA for Canonical Auction Games.

Trading Agents lecture published

Morgan & Claypool has just published my volume Trading Agents, in its “Synthesis Lecture” series. Check out the description here.
, ,


The 2010 AAAI Conference was held last week in Atlanta.  SRG presented one paper there: Algorithms for Finding Approximate Formations in Games (PR Jordan & MPW) The conference also featured papers co-authored by several SRG alumni Automated…

2010 Trading Agent Competition Results

The eleventh annual Trading Agent Competition was completed earlier this month at the ACM EC-10 conference in Cambridge, Massachusetts.  TAC-10 featured three games: the supply chain management (SCM), market design (reverse TAC, or "CAT"),…

Autonomous Bidding Agents: Strategies and Lessons from the Trading Agent Competition

E-commerce increasingly provides opportunities for autonomous bidding agents: computer programs that bid in electronic markets without direct human intervention. Automated bidding strategies for an auction of a single good with a known valuation are fairly straightforward; designing strategies for simultaneous auctions with interdependent valuations is a more complex undertaking. This book presents algorithmic advances and strategy ideas within an integrated bidding agent architecture that have emerged from recent work in this fast-growing area of research in academia and industry. The authors analyze several novel bidding approaches that developed from the Trading Agent Competition (TAC), held annually since 2000. The benchmark challenge for competing agents—to buy and sell multiple goods with interdependent valuations in simultaneous auctions of different types—encourages competitors to apply innovative techniques to a common task. The book traces the evolution of TAC and follows selected agents from conception through several competitions, presenting and analyzing detailed algorithms developed for autonomous bidding. Autonomous Bidding Agents provides the first integrated treatment of methods in this rapidly developing domain of AI. The authors—who introduced TAC and created some of its most successful agents—offer both an overview of current research and new results.

Threads of Research on Generic CDA Strategies

Research on trading strategy for generic continuous double auctions (CDAs) seems to take place on four parallel and minimally interacting threads. By “generic CDA”, I mean models of two-sided continuous trading of an abstract good, as distinct…

Technology Enablers of Latency Arbitrage

Ralph Frankel, CTO of Solace Systems, has a fascinating article on the technology for shaving microseconds and milliseconds off the market-response time they can achieve for high-frequency trading functions. He classifies “tricks of the trade”…

Short-Lived Dark Pools

Felix Salmon cited my original post on employing one-second call markets as a counter to high-frequency trading. He ends his post by raising the following question for his readers (far more numerous than mine) to consider: Would this plan essentially…

Google Faculty Summit

I just returned from the Google Faculty Summit, a gathering of ~100 professors (mostly computer scientists) in Mountain View from universities across North America, and a few from South America as well. Google holds this event annually, as part…

AAAI Asilomar Meeting

John Markoff’s NYT article “Scientists Worry Machines May Outsmart Man” touched off a mini-firestorm this week. The article refers to a meeting of AI scientists held at Asilomar (a conference center near Monterey) in February to discuss…

Cost/Benefit of High-Frequency Trading

On Marginal Revolution, Tyler Cowen discusses high-frequency trading and gets to the nub of the issue: The philosophical question is why it might possibly be beneficial to have market prices adjust within five seconds rather than within fifteen.…

Countering High-Frequency Trading

The recent NYT article by Charles Duhigg on high-frequency trading (HFT) has set off a flurry of argument about the benefits and threats of this activity to financial trading systems. The revelation that some systems provide advance information…

2009 Trading Agent Competition Results

The tenth annual Trading Agent Competition was completed earlier this month at the IJCAI-09 conference in Pasadena, California. TAC-09 featured three games: the supply chain management, market design, and ad auction games. Preliminary rounds began in June, involving 42 teams from 14 countries.