Incentive mechanisms for crowdsourcing are designed to incentivize financially self-interested workers to generate and report high-quality labels. Existing mechanisms are often developed as one-shot static solutions, assuming a certain level of knowledge about worker models (expertise levels, costs for exerting efforts, etc.). In this paper, we propose a novel inference aided reinforcement mechanism that acquires data sequentially and requires no such prior assumptions. Specifically, we first design a Gibbs sampling augmented Bayesian inference algorithm to estimate workers' labeling strategies from the collected labels at each step. Then we propose a reinforcement incentive learning (RIL) method, building on top of the above estimates, to uncover how workers respond to different payments. RIL dynamically determines the payment without accessing any ground-truth labels. We theoretically prove that RIL is able to incentivize rational workers to provide high-quality labels both at each step and in the long run. Empirical results show that our mechanism performs consistently well under both rational and non-fully rational (adaptive learning) worker models. Besides, the payments offered by RIL are more robust and have lower variances compared to existing one-shot mechanisms.

Incentive mechanisms for crowdsourcing are designed to incentivize financially self-interested workers to generate and report high-quality labels. Existing mechanisms are often developed as one-shot static solutions, assuming a certain level of knowledge about worker models (expertise levels, costs for exerting efforts, etc.). In this paper, we propose a novel inference aided reinforcement mechanism that acquires data sequentially and requires no such prior assumptions. Specifically, we first design a Gibbs sampling augmented Bayesian inference algorithm to estimate workers' labeling strategies from the collected labels at each step. Then we propose a reinforcement incentive learning (RIL) method, building on top of the above estimates, to uncover how workers respond to different payments. RIL dynamically determines the payment without accessing any ground-truth labels. We theoretically prove that RIL is able to incentivize rational workers to provide high-quality labels both at each step and in the long run. Empirical results show that our mechanism performs consistently well under both rational and non-fully rational (adaptive learning) worker models. Besides, the payments offered by RIL are more robust and have lower variances compared to existing one-shot mechanisms.

We theoretically prove that RIL is able to incentivize rational workers to provide high-quality labels.

Neural Information Processing Systems http://nips.cc/

The recent success of denoising diffusion models has significantly advanced text-to-image generation. While these large-scale pretrained models show excellent performance in general image synthesis, downstream objectives often require fine-tuning to meet specific criteria such as aesthetics or human preference. Reward gradient-based strategies are promising in this context, yet existing methods are limited to single-reward tasks, restricting their applicability in real-world scenarios that demand adapting to multiple objectives introduced incrementally over time. In this paper, we first define this more realistic and unexplored problem, termed Reward Incremental Learning (RIL), where models are desired to adapt to multiple downstream objectives incrementally. Additionally, while the models adapt to the ever-emerging new objectives, we observe a unique form of catastrophic forgetting in diffusion model fine-tuning, affecting both metric-wise and visual structure-wise image quality. To address this catastrophic forgetting challenge, we propose Reward Incremental Distillation (RID), a method that mitigates forgetting with minimal computational overhead, enabling stable performance across sequential reward tasks. The experimental results demonstrate the efficacy of RID in achieving consistent, high-quality generation in RIL scenarios. The source code of our work will be publicly available upon acceptance.

In recent years, persistent homology, one of the flagship concepts of Topological Data Analysis (TDA), has found its use in many fields such as neuroscience, material science, sensor networks, shape recognition, gene expression data analysis and many more Giunti et al. (2022). The performance of machine learning models such as Graph Neural Networks (GNNs) can be enhanced by augmenting topological information captured by persistent homology (Hofer et al., 2017; Dehmamy et al., 2019; Carrière et al., 2020; Horn et al., 2022). Classical persistent homology, also known as 1-parameter persistence, captures the evolution of topological structures in a simplicial complex K following a filter function f: K R. The evolution of topological structures, in this case, can be completely characterized and compactly represented as persistence diagrams or equivalently barcodes Zomorodian & Carlsson (2004); Edelsbrunner et al. (2002). These persistence diagrams or barcodes can be vectorized Bubenik (2015); Reininghaus et al. (2015); Adams et al. (2017); Hofer et al. (2019); Carrière et al. (2020); Kim et al. (2020) and used in machine learning pipelines. In most applications, the simplicial complex K is given and the choice of the filter function f depends on the application. Choosing an appropriate filter function can be challenging. To avoid this, in Hofer et al. (2020), the authors proposed an end-to-end learning framework to learn the filter function rather than relying on making the right choice. They showed that learning the filter function performs better than the standard choices of filter functions on many graph datasets.

$1$-parameter persistent homology, a cornerstone in Topological Data Analysis (TDA), studies the evolution of topological features such as connected components and cycles hidden in data. It has been applied to enhance the representation power of deep learning models, such as Graph Neural Networks (GNNs). To enrich the representations of topological features, here we propose to study $2$-parameter persistence modules induced by bi-filtration functions. In order to incorporate these representations into machine learning models, we introduce a novel vector representation called Generalized Rank Invariant Landscape (GRIL) for $2$-parameter persistence modules. We show that this vector representation is $1$-Lipschitz stable and differentiable with respect to underlying filtration functions and can be easily integrated into machine learning models to augment encoding topological features. We present an algorithm to compute the vector representation efficiently. We also test our methods on synthetic and benchmark graph datasets, and compare the results with previous vector representations of $1$-parameter and $2$-parameter persistence modules. Further, we augment GNNs with GRIL features and observe an increase in performance indicating that GRIL can capture additional features enriching GNNs. We make the complete code for the proposed method available at https://github.com/soham0209/mpml-graph.

We accept proposals only via our submission system ExOrdo. And if your sweetheart theme isn't listed, dare yet to send your proposal under the category "Other". Don't let the limits of our imagination limit yourself! If your proposal gets accepted, we will ask you to submit your presentation material to be shared on the event platform (shareable version). Please note that we also intend to record all the sessions and recordings will be shared on the event platform.

The high interest towards the first truly global Artificial Intelligence conference in Architecture, Engineering and Construction industry in 2021 gave the spark for the annual conferences. In 2021 event, there were 235 registered participants from 30 countries, who graded the overall conference experience over 4 on a scale 1-5. The 2021 conference was already successful, but in 2022 we are heading to have even bigger and more international event. How society and economy are changing through digital disruption is one of the defining issues of our time. Throughout the history of the mankind, we have had a quest for seeking solutions for better human life.

Incentive mechanisms for crowdsourcing are designed to incentivize financially self-interested workers to generate and report high-quality labels. Existing mechanisms are often developed as one-shot static solutions, assuming a certain level of knowledge about worker models (expertise levels, costs for exerting efforts, etc.). In this paper, we propose a novel inference aided reinforcement mechanism that acquires data sequentially and requires no such prior assumptions. Specifically, we first design a Gibbs sampling augmented Bayesian inference algorithm to estimate workers' labeling strategies from the collected labels at each step. Then we propose a reinforcement incentive learning (RIL) method, building on top of the above estimates, to uncover how workers respond to different payments.