We study the problem of forecasting the number of units fulfilled (or ``drained'') from each inventory warehouse to meet customer demand, along with the associated outbound shipping costs. The actual drain and shipping costs are determined by complex production systems that manage the planning and execution of customers' orders fulfillment, i.e. from where and how to ship a unit to be delivered to a customer. Accurately modeling these processes is critical for regional inventory planning, especially when using Reinforcement Learning (RL) to develop control policies. For the RL usecase, a drain model is incorporated into a simulator to produce long rollouts, which we desire to be differentiable. While simulating the calls to the internal software systems can be used to recover this transition, they are non-differentiable and too slow and costly to run within an RL training environment. Accordingly, we frame this as a probabilistic forecasting problem, modeling the joint distribution of outbound drain and shipping costs across all warehouses at each time period, conditioned on inventory positions and exogenous customer demand. To ensure robustness in an RL environment, the model must handle out-of-distribution scenarios that arise from off-policy trajectories. We propose a validation scheme that leverages production systems to evaluate the drain model on counterfactual inventory states induced by RL policies. Preliminary results demonstrate the model's accuracy within the in-distribution setting.

As robots become increasingly involved in decision-making processes (e.g., personnel selection), concerns about fairness and social inclusion arise. This study examines social exclusion in robot-led group interviews by robot Ameca, exploring the relationship between objective exclusion (robot's attention allocation), subjective exclusion (perceived exclusion), mood change, and need fulfillment. In a controlled lab study (N = 35), higher objective exclusion significantly predicted subjective exclusion. In turn, subjective exclusion negatively impacted mood and need fulfillment but only mediated the relationship between objective exclusion and need fulfillment. A piecewise regression analysis identified a critical threshold at which objective exclusion begins to be perceived as subjective exclusion. Additionally, the standing position was the primary predictor of exclusion, whereas demographic factors (e.g., gender, height) had no significant effect. These findings underscore the need to consider both objective and subjective exclusion in human-robot interactions and have implications for fairness in robot-assisted hiring processes.

Practitioners designing reinforcement learning policies face a fundamental challenge: translating intended behavioral objectives into representative reward functions. This challenge stems from behavioral intent requiring simultaneous achievement of multiple competing objectives, typically addressed through labor-intensive linear reward composition that yields brittle results. Consider the ubiquitous robotics scenario where performance maximization directly conflicts with energy conservation. Such competitive dynamics are resistant to simple linear reward combinations. In this paper, we present the concept of objective fulfillment upon which we build Fulfillment Priority Logic (FPL). FPL allows practitioners to define logical formula representing their intentions and priorities within multi-objective reinforcement learning. Our novel Balanced Policy Gradient algorithm leverages FPL specifications to achieve up to 500\% better sample efficiency compared to Soft Actor Critic. Notably, this work constitutes the first implementation of non-linear utility scalarization design, specifically for continuous control problems.

The handover (HO) procedure is one of the most critical functions in a cellular network driven by measurements of the user channel of the serving and neighboring cells. The success rate of the entire HO procedure is significantly affected by the preparation stage. As massive Multiple-Input Multiple-Output (MIMO) systems with large antenna arrays allow resolving finer details of channel behavior, we investigate how machine learning can be applied to time series data of beam measurements in the Fifth Generation (5G) New Radio (NR) system to improve the HO procedure. This paper introduces the Early-Scheduled Handover Preparation scheme designed to enhance the robustness and efficiency of the HO procedure, particularly in scenarios involving high mobility and dense small cell deployments. Early-Scheduled Handover Preparation focuses on optimizing the timing of the HO preparation phase by leveraging machine learning techniques to predict the earliest possible trigger points for HO events. We identify a new early trigger for HO preparation and demonstrate how it can beneficially reduce the required time for HO execution reducing channel quality degradation. These insights enable a new HO preparation scheme that offers a novel, user-aware, and proactive HO decision making in MIMO scenarios incorporating mobility.

Conditional question answering (CQA) is an important task that aims to find probable answers and identify conditions that need to be satisfied to support the answer. Existing approaches struggle with CQA due to two main challenges: (1) precisely identifying conditions and their logical relationship, and (2) verifying and solving the conditions. To address these challenges, we propose Chain of Condition, a novel prompting approach by firstly identifying all conditions and constructing their logical relationships explicitly according to the document, then verifying whether these conditions are satisfied, finally solving the logical expression by tools to indicate any missing conditions and generating the answer based on the resolved conditions. The experiments on two benchmark conditional question answering datasets shows chain of condition outperforms existing prompting baselines, establishing a new state-of-the-art. Furthermore, with backbone models like GPT-3.5-Turbo or GPT-4, it surpasses all supervised baselines with only few-shot settings.

The importance of these seed stocks is underscored by the critical need for timely fulfillment of seed orders to meet specific planting windows, often mandated by the seasonal growth cycles of different crops. Failure to meet these strict timelines can lead to a host of downstream issues, including suboptimal crop yields and financial loss [1]. Figure 1: Overview of the centralized seed fulfillment process. The process begins with the arrival of seed stocks from multiple sites with stochastic, a priori unknown arrival distributions and ends with the fulfillment of orders with different deadlines and quantities. Our proposed adaptive adaptive hybrid tree search approach provides an efficient solution to the wave scheduling problem, optimizing the process of order fulfillment. Order fulfillment in industries such as e-commerce [2] and retail [3] often involve centralized fulfillment centers that simultaneously process arriving inventory and fulfill orders based on their deadlines. The fulfillment process with large catalogs often handle a batch of orders, hereinafter referred to as wave, together using automated sortation systems [4]. The supply chain in these sectors is typically well-established, with known inventory quantities and deterministic restock times. The problem of optimally scheduling waves to maximize fulfillment efficiency is addressed using traditional operations research and optimization techniques [5], [6] as order deadlines and inventory levels are known a priori or can be forecasted with low uncertainty.

Simulating a single trajectory of a dynamical system under some state-dependent policy is a core bottleneck in policy optimization algorithms. The many inherently serial policy evaluations that must be performed in a single simulation constitute the bulk of this bottleneck. To wit, in applying policy optimization to supply chain optimization (SCO) problems, simulating a single month of a supply chain can take several hours. We present an iterative algorithm for policy simulation, which we dub Picard Iteration. This scheme carefully assigns policy evaluation tasks to independent processes. Within an iteration, a single process evaluates the policy only on its assigned tasks while assuming a certain 'cached' evaluation for other tasks; the cache is updated at the end of the iteration. Implemented on GPUs, this scheme admits batched evaluation of the policy on a single trajectory. We prove that the structure afforded by many SCO problems allows convergence in a small number of iterations, independent of the horizon. We demonstrate practical speedups of 400x on large-scale SCO problems even with a single GPU, and also demonstrate practical efficacy in other RL environments.

Intent Management Function (IMF) is an integral part of future-generation networks. In recent years, there has been some work on AI-based IMFs that can handle conflicting intents and prioritize the global objective based on apriori definition of the utility function and accorded priorities for competing intents. Some of the earlier works use Multi-Agent Reinforcement Learning (MARL) techniques with AdHoc Teaming (AHT) approaches for efficient conflict handling in IMF. However, the success of such frameworks in real-life scenarios requires them to be flexible to business situations. The intent priorities can change and the utility function, which measures the extent of intent fulfilment, may also vary in definition. This paper proposes a novel mechanism whereby the IMF can generalize to different forms of utility functions and change of intent priorities at run-time without additional training. Such generalization ability, without additional training requirements, would help to deploy IMF in live networks where customer intents and priorities change frequently. Results on the network emulator demonstrate the efficacy of the approach, scalability for new intents, outperforming existing techniques that require additional training to achieve the same degree of flexibility thereby saving cost, and increasing efficiency and adaptability.

Counterfactual (CF) explanations, also known as contrastive explanations and algorithmic recourses, are popular for explaining machine learning models in high-stakes domains. For a subject that receives a negative model prediction (e.g., mortgage application denial), the CF explanations are similar instances but with positive predictions, which informs the subject of ways to improve. While their various properties have been studied, such as validity and stability, we contribute a novel one: their behaviors under iterative partial fulfillment (IPF). Specifically, upon receiving a CF explanation, the subject may only partially fulfill it before requesting a new prediction with a new explanation, and repeat until the prediction is positive. Such partial fulfillment could be due to the subject's limited capability (e.g., can only pay down two out of four credit card accounts at this moment) or an attempt to take the chance (e.g., betting that a monthly salary increase of \$800 is enough even though \$1,000 is recommended). Does such iterative partial fulfillment increase or decrease the total cost of improvement incurred by the subject? We mathematically formalize IPF and demonstrate, both theoretically and empirically, that different CF algorithms exhibit vastly different behaviors under IPF. We discuss implications of our observations, advocate for this factor to be carefully considered in the development and study of CF algorithms, and give several directions for future work.

AI technology is quickly creeping into every industry, prompting new questions about whether online content comes from a human or a computer. Hallow app CEO and co-creator Alex Jones says that while artificial intelligence (AI) can be used for both evil and good, there is one central question people must ask when it comes to emerging technologies. Jones said that a lot of friends and "fellow startup folks" who are working in AI are building "really mind-blowing tools that can do a lot of different things." AI is considered by many as one of the "scarier technologies," which can be used for "tremendous, tremendous evil," said Jones, naming the the internet's "massive proliferation of pornography" as just one example. AI GIVES GOOGLE POWER TO'DICTATE' THE NEWS PEOPLE SEE, WHAT THEY BUY, HOW THEY VOTE, ATTORNEY CLAIMS Technology can also be used for good, he continued, including advances in medical care and of connecting long-lost loved ones by acting as a vehicle "to allow God to reach into people's lives."