The 3rd Workshop on Causal Inference and Machine Learning in Practice at KDD 2025 aims to bring together researchers, industry professionals, and practitioners to explore the application of causal inference within machine learning models. As causal machine learning techniques gain traction across industries, practical challenges related to trustworthiness, robustness, and fairness remain at the forefront. This workshop will provide a forum to discuss methodologies for evaluating causal models in real-world scenarios and explore innovative applications that integrate causal inference with generative AI (GenAI) and large language models (LLMs). Topics of interest include using GenAI and LLMs to facilitate causal inference tasks and leveraging causal inference techniques for evaluating and improving GenAI/LLM models. Building on the success of the previous workshop editions at KDD 2023 and KDD 2024, which attracted over 200 and 250 participants, respectively, this workshop will continue fostering collaboration between academia and industry. Through invited talks, contributed papers, and interactive discussions, we will address key challenges and opportunities at the intersection of causal inference and machine learning. As the field continues to evolve, this workshop serves as a crucial platform for knowledge exchange and innovation, driving forward the application of causal techniques in machine learning and AI.

The workshop's rationale stems from the escalating interest in causal inference and machine learning methodologies within various industrial contexts. This surge in demand underscores the importance for both scholars and practitioners to exchange knowledge and best practices regarding the application of these techniques to tackle real-world challenges. Yet, applying causal machine learning techniques in real-world scenarios presents a range of challenges not addressed in the academic literature. This workshop aims to address the challenges for practical causal machine learning and explore new industry use cases. The workshop will provide a forum for practitioners and researchers to exchange ideas and explore new collaborations. Moreover, this workshop aims to capitalize on the success and achievements of the KDD 2023 Workshop titled "Causal Inference and Machine Learning in Practice".

The increasing demand for data-driven decision-making has led to the rapid growth of machine learning applications in various industries. However, the ability to draw causal inferences from observational data remains a crucial challenge. In recent years, causal inference has emerged as a powerful tool for understanding the effects of interventions in complex systems. Combining causal inference with machine learning has the potential to provide a deeper understanding of the underlying mechanisms and to develop more effective solutions to real-world problems.

In recent years, both academic research and industry applications see an increased effort in using machine learning methods to measure granular causal effects and design optimal policies based on these causal estimates. Open source packages such as CausalML and EconML provide a unified interface for applied researchers and industry practitioners with a variety of machine learning methods for causal inference. The tutorial will cover the topics including conditional treatment effect estimators by meta-learners and tree-based algorithms, model validations and sensitivity analysis, optimization algorithms including policy leaner and cost optimization. In addition, the tutorial will demonstrate the production of these algorithms in industry use cases.

This panel aims to address areas that are widely acknowledged to be of critical importance to the success of Data Science projects and to the healthy growth of KDD/Data Science as a field of scientific research. However, despite this acknowledgement of their criticality, these areas receive insufficient attention in the major conferences in the field. Furthermore, there is a lack of actual actions and tools to address these areas in actual practice. These areas are summarized as follows:

Although scheduling multiple tasks in motor learning to maximize long-term retention of performance is of great practical importance in sports training and motor rehabilitation after brain injury, it is unclear how to do so. We propose here a novel theoretical approach that uses optimal control theory and computational models of motor adaptation to determine schedules that maximize long-term retention predictively. Using Pontryagin's maximum principle, we derived a control law that determines the trial-by-trial task choice that maximizes overall delayed retention for all tasks, as predicted by the state-space model. Simulations of a single session of adaptation with two tasks show that when task interference is high, there exists a threshold in relative task difficulty below which the alternating schedule is optimal. Only for large differences in task difficulties do optimal schedules assign more trials to the harder task. However, over the parameter range tested, alternating schedules yield long-term retention performance that is only slightly inferior to performance given by the true optimal schedules. Our results thus predict that in a large number of learning situations wherein tasks interfere, intermixing tasks with an equal number of trials is an effective strategy in enhancing long-term retention.

Being readily available in most of 802.11 radios, multirate capability appears to be useful as WiFi networks are getting more prevalent and crowded. More specifically, it would be helpful in high-density scenarios because internode distance is short enough to employ high data rates. However, communication at high data rates mandates a large number of hops for a given node pair in a multihop network and thus, can easily be depreciated as per-hop overhead at several layers of network protocol is aggregated over the increased number of hops. This paper presents a novel multihop, multirate adaptation mechanism, called multihop transmission opportunity (MTOP), that allows a frame to be forwarded a number of hops consecutively to minimize the MAC-layer overhead between hops. This seemingly collision-prone nonstop forwarding is proved to be safe via analysis and USRP/GNU Radio-based experiment in this paper. The idea of MTOP is in clear contrast to the conventional opportunistic transmission mechanism, known as TXOP, where a node transmits multiple frames back-to-back when it gets an opportunity in a single-hop WLAN. We conducted an extensive simulation study via OPNET, demonstrating the performance advantage of MTOP under a wide range of network scenarios.

Wireless multihop communication is becoming more important due to the increasing popularity of wireless sensor networks, wireless mesh networks, and mobile social networks. They are distinguished from conventional multihop networks in terms of scale, traffic intensity and/or node density. Being readily-available in most of 802.11 radios, multirate facility appears to be useful to address some of these issues and is particularly helpful in high-density scenarios where inter-node distance is short, demanding a prudent multirate adaptation algorithm. However, communication at high bit rates mandates a large number of hops for a given node pair and thus, can easily be depreciated as per-hop overhead at several layers of network protocol is aggregated over the increased number of hops. This paper presents a novel multihop, multirate adaptation mechanism, called Multihop Transmission OPportunity (MTOP), that allows a frame to be forwarded a number of hops consecutively but reduces the MAC-layer overhead between hops. This seemingly collision-prone multihop forwarding is proven to be safe via analysis and USRP/GNU Radio-based experiment. The idea of MTOP is in clear contrast to, but not mutually exclusive with, the conventional opportunistic transmission mechanism, referred to as TXOP, where a node transmits multiple frames back-to-back when it gets an opportunity. We conducted an extensive simulation study via ns-2, demonstrating the performance advantage of MTOP under a wide range of network scenarios.

Wireless network coding (WNC) has been considered as a promising solution with its capability of improving the overall performance of wireless mesh networks. A problem caused when WNC is employed in multi-rate wireless networks is that the link status between the relay node and intended receivers of a coded packet is mutually different. Even though there are links that support higher transmission rates, the relay node must transmit the coded packet at the lowest transmission rate among intended links' rates for the successful reception at all intended receivers. In this paper, we propose a WNC scheme that is combined with Forward Error Correction (FEC) to alleviate the problem explained above. The proposed scheme tries to transmit coded packets with a transmission rate higher than the lowest transmission rate so that the time required to transmit coded packets is reduced. However, it may increase the error rate. To compensate for the increased error rate, we use adaptively controlled FEC code. Our simulation study results show that the proposed scheme can increase the network coding gain compared to existing schemes.