ParetoFlow: Guided Flows in Multi-Objective Optimization
Authors: Ye Yuan, Can Chen, Christopher Pal, Xue Liu
ICLR 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We conduct comprehensive experiments to evaluate our method. In Section 4.4, we compare our approach to several baselines to assess performance. In Section 4.5, we demonstrate the effectiveness of our proposed modules. |
| Researcher Affiliation | Academia | 1Mc Gill, 2MILA Quebec AI Institute, 3Polytechnique Montreal, 4Canada CIFAR AI Chair |
| Pseudocode | Yes | Algorithm 1 Pareto Flow: Guided Flows in Multi-Objective Optimization |
| Open Source Code | No | Our code will be available here. |
| Open Datasets | Yes | We utilize the Off-MOO-Bench, which summarizes and collects several established benchmarks Xue et al. (2024). We explore five groups of tasks, each task with a dataset D and a groundtruth oracle f for evaluation, which is not queried during training. For discrete inputs, we convert them to continuous logits as suggested by Trabucco et al. (2022); Xue et al. (2024). |
| Dataset Splits | No | The paper does not explicitly provide training/test/validation dataset splits for the offline dataset D itself, nor for its models. It refers to an 'offline dataset D' and a 'groundtruth oracle f for evaluation' but lacks specific partitioning details. |
| Hardware Specification | Yes | All experiments are conducted on a workstation with an Intel i9-12900K CPU and an NVIDIA RTX3090 GPU. |
| Software Dependencies | No | The paper mentions using 'Adam optimizer Kingma & Ba (2015)' and refers to MLP architectures but does not specify software library versions (e.g., PyTorch, TensorFlow, or Python versions) that would be needed for replication. |
| Experiment Setup | Yes | Each predictor consists of a 3-layer MLP with ReLU activations, featuring a hidden layer size of 2048. These models are trained over 200 epochs with a batch size of 128, using the Adam optimizer Kingma & Ba (2015) at an initial learning rate of 1e-3, and a decay rate of 0.98 per epoch. Flow matching training follows protocols from Tomczak (2022), employing a 4-layer MLP with SeLU activations and a hidden layer size of 512. The model undergoes 1000 training epochs with early stopping after 20 epochs of no improvement, with a batch size of 128 and the Adam optimizer. We set the number of neighboring distributions, K, to be m + 1, where m is the number of objective functions, and set the number of offspring, O, to be 5. |