Managing Disruption Risks in Multimodal Transportation Systems
The rapid growth of the freight transport industry raises concerns about its environmental impact, particularly its reliance on fossil fuels. Containerized intermodal transport offers a promising solution; however, its fast-paced nature and reliance on multiple transport modes make it vulnerable to disruptions. Even minor issues can cause delays, cancellations, and reduced efficiency. To ensure operational continuity, stakeholders—including shippers, carriers, and Logistics Service Providers—should adopt risk mitigation strategies. This talk examines strategies for managing disruptions and enhancing sustainability to improve system performance.
Présentatrice :
- Asefeh Hasani, Operations Research and AI Scientist at Termont
LT-Soups: Bridging Head and Tail Classes via Subsampled Model Soups
Real-world datasets typically exhibit long-tailed (LT) distributions, where a few head classes dominate, and many tail classes are severely underrepresented. While recent work shows that parameter-efficient fine-tuning (PEFT) methods like LoRA and AdaptFormer preserve tail-class performance on foundation models such as CLIP, we find that they do so at the cost of head-class accuracy. We identify the head-tail ratio, the proportion of head to tail classes, as a crucial but overlooked factor influencing this trade-off. Through controlled experiments on CIFAR100 with varying imbalance and head-tail ratios, we show that PEFT excels in tail-heavy scenarios but degrades in more balanced and head-heavy distributions. To overcome these limitations, we propose LT-Soups, a two-stage model soups framework designed to generalize across diverse LT regimes. In the first stage, LT-Soups averages models fine-tuned on balanced subsets to reduce head-class bias; in the second, it fine-tunes only the classifier on the full dataset to restore head-class accuracy. Experiments across six benchmark datasets show that LT-Soups achieves superior trade-offs compared to both PEFT and traditional model soups across a wide range of imbalance regimes.
Présentateur :
- Masih Aminbeidokhti, Ph. D student, LIVIA and ILLS
Renseignements :
Pizza offerte. Le séminaire sera présenté en anglais.