1 code implementation • 12 Mar 2024 • Huijie Tang, Federico Berto, Jinkyoo Park
We first propose a selective communication block to gather richer information for better agent coordination within multi-agent environments and train the model with a Q-learning-based algorithm.
Multi-Agent Path Finding Multi-agent Reinforcement Learning +1
2 code implementations • 11 Mar 2024 • Minsu Kim, Sanghyeok Choi, Jiwoo Son, Hyeonah Kim, Jinkyoo Park, Yoshua Bengio
This paper introduces the Generative Flow Ant Colony Sampler (GFACS), a novel neural-guided meta-heuristic algorithm for combinatorial optimization.
no code implementations • 28 Feb 2024 • Shiqi Lei, Kanghoon Lee, Linjing Li, Jinkyoo Park, Jiachen Li
Offline learning has become widely used due to its ability to derive effective policies from offline datasets gathered by expert demonstrators without interacting with the environment directly.
1 code implementation • 23 Feb 2024 • Huijie Tang, Federico Berto, Zihan Ma, Chuanbo Hua, Kyuree Ahn, Jinkyoo Park
With a simple training scheme and implementation, HiMAP demonstrates competitive results in terms of success rate and scalability in the field of imitation-learning-only MAPF, showing the potential of imitation-learning-only MAPF equipped with inference techniques.
no code implementations • 8 Feb 2024 • Nayoung Kim, Minsu Kim, Jinkyoo Park
Antibody design plays a pivotal role in advancing therapeutics.
no code implementations • 5 Feb 2024 • Hyeonah Kim, Minsu Kim, Sanghyeok Choi, Jinkyoo Park
This paper proposes a novel variant of GFlowNet, genetic-guided GFlowNet (Genetic GFN), which integrates an iterative genetic search into GFlowNet.
no code implementations • 22 Jan 2024 • Jiachen Li, Chuanbo Hua, Hengbo Ma, Jinkyoo Park, Victoria Dax, Mykel J. Kochenderfer
In this paper, we propose a systematic relational reasoning approach with explicit inference of the underlying dynamically evolving relational structures, and we demonstrate its effectiveness for multi-agent trajectory prediction and social robot navigation.
no code implementations • 27 Nov 2023 • Jiachen Li, David Isele, Kanghoon Lee, Jinkyoo Park, Kikuo Fujimura, Mykel J. Kochenderfer
Moreover, we propose an interactivity estimation mechanism based on the difference between predicted trajectories in these two situations, which indicates the degree of influence of the ego agent on other agents.
no code implementations • 22 Oct 2023 • Abhay Sobhanan, Junyoung Park, Jinkyoo Park, Changhyun Kwon
For each higher-level decision candidate, we predict the objective function values of the underlying vehicle routing problems using a pre-trained graph neural network without actually solving the routing problems.
1 code implementation • 4 Oct 2023 • Minsu Kim, Joohwan Ko, Taeyoung Yun, Dinghuai Zhang, Ling Pan, Woochang Kim, Jinkyoo Park, Emmanuel Bengio, Yoshua Bengio
We find that the challenge is greatly reduced if a learned function of the temperature is used to scale the policy's logits directly.
2 code implementations • 4 Oct 2023 • Minsu Kim, Taeyoung Yun, Emmanuel Bengio, Dinghuai Zhang, Yoshua Bengio, Sungsoo Ahn, Jinkyoo Park
Generative Flow Networks (GFlowNets) are amortized sampling methods that learn a distribution over discrete objects proportional to their rewards.
no code implementations • 19 Jul 2023 • Kanghoon Lee, Jiachen Li, David Isele, Jinkyoo Park, Kikuo Fujimura, Mykel J. Kochenderfer
Although deep reinforcement learning (DRL) has shown promising results for autonomous navigation in interactive traffic scenarios, existing work typically adopts a fixed behavior policy to control social vehicles in the training environment.
1 code implementation • 29 Jun 2023 • Hyeonah Kim, Jinkyoo Park, Changhyun Kwon
We design a learning-based separation heuristic algorithm with graph coarsening that learns the solutions of the exact separation problem with a graph neural network (GNN), which is trained with small instances of 50 to 100 customers.
1 code implementation • 29 Jun 2023 • Federico Berto, Chuanbo Hua, Junyoung Park, Minsu Kim, Hyeonah Kim, Jiwoo Son, Haeyeon Kim, Joungho Kim, Jinkyoo Park
To address these challenges, we introduce RL4CO, a unified Reinforcement Learning (RL) for Combinatorial Optimization (CO) library.
1 code implementation • NeurIPS 2023 • Minsu Kim, Federico Berto, Sungsoo Ahn, Jinkyoo Park
The subsequent stage involves bootstrapping, which augments the training dataset with self-generated data labeled by a proxy score function.
1 code implementation • 5 Jun 2023 • Jiwoo Son, Minsu Kim, Sanghyeok Choi, Hyeonah Kim, Jinkyoo Park
Notably, our method achieves significant reductions of runtime, approximately 335 times, and cost values of about 53\% compared to a competitive heuristic (LKH3) in the case of 100 vehicles with 1, 000 cities of mTSP.
1 code implementation • 5 Jun 2023 • Jiwoo Son, Minsu Kim, Hyeonah Kim, Jinkyoo Park
First, SML transforms the context embedding for subsequent adaptation of SAGE based on scale information.
no code implementations • 2 Jun 2023 • Hyeonah Kim, Minsu Kim, Sungsoo Ahn, Jinkyoo Park
Deep reinforcement learning (DRL) has significantly advanced the field of combinatorial optimization (CO).
no code implementations • 5 Feb 2023 • Vivian W. H. Wong, Sang Hun Kim, Junyoung Park, Jinkyoo Park, Kincho H. Law
The interrupting swap-allowed blocking job shop problem (ISBJSSP) is a complex scheduling problem that is able to model many manufacturing planning and logistics applications realistically by addressing both the lack of storage capacity and unforeseen production interruptions.
no code implementations • 22 Nov 2022 • Haewon Jung, Junyoung Park, Jinkyoo Park
Convex quadratic programming (QP) is an important sub-field of mathematical optimization.
no code implementations • 22 Oct 2022 • Yohan Jung, Jinkyoo Park
To remedy this issue, we introduce Bayesian convolutional deep sets that construct the random translation equivariant functional representations with stationary prior.
no code implementations • 10 Aug 2022 • Jiachen Li, Chuanbo Hua, Jinkyoo Park, Hengbo Ma, Victoria Dax, Mykel J. Kochenderfer
While the modeling of pair-wise relations has been widely studied in multi-agent interacting systems, its ability to capture higher-level and larger-scale group-wise activities is limited.
no code implementations • 6 Jun 2022 • Minjun Kim, Junyoung Park, Jinkyoo Park
Inspired by CE, we propose Neuro CE (NCE), a fundamental operator of learned meta-heuristic, to solve various VRPs while overcoming the limitations of CE (i. e., the expensive $\mathcal{O}(n^4)$ search cost).
no code implementations • 1 Jun 2022 • Junyoung Park, Federico Berto, Arec Jamgochian, Mykel J. Kochenderfer, Jinkyoo Park
In this paper, we propose Meta-SysId, a meta-learning approach to model sets of systems that have behavior governed by common but unknown laws and that differentiate themselves by their context.
2 code implementations • 26 May 2022 • Haeyeon Kim, Minsu Kim, Federico Berto, Joungho Kim, Jinkyoo Park
In this paper, we present DevFormer, a novel transformer-based architecture for addressing the complex and computationally demanding problem of hardware design optimization.
1 code implementation • 26 May 2022 • Minsu Kim, Junyoung Park, Jinkyoo Park
Deep reinforcement learning (DRL)-based combinatorial optimization (CO) methods (i. e., DRL-NCO) have shown significant merit over the conventional CO solvers as DRL-NCO is capable of learning CO solvers less relying on problem-specific expert domain knowledge (heuristic method) and supervised labeled data (supervised learning method).
1 code implementation • NeurIPS Workshop DLDE 2021 • Federico Berto, Stefano Massaroli, Michael Poli, Jinkyoo Park
Synthesizing optimal controllers for dynamical systems often involves solving optimization problems with hard real-time constraints.
1 code implementation • NeurIPS 2021 • Minsu Kim, Jinkyoo Park, Joungho Kim
Recently, deep reinforcement learning (DRL) frameworks have shown potential for solving NP-hard routing problems such as the traveling salesman problem (TSP) without problem-specific expert knowledge.
no code implementations • 29 Sep 2021 • Junyoung Park, Chihyeon Song, Jinkyoo Park
On the physical heat diffusion, we further apply ICGNN to solve a design optimization problem, which seeks to find the optimal heater allocations while considering the optimal operation of the heaters, by using a gradient-based method.
no code implementations • 29 Sep 2021 • Junyoung Park, Fangying Chen, Jinkyoo Park
We show that our model is able to outperform other baseline methods for most of the datasets.
no code implementations • 29 Sep 2021 • Kyuree Ahn, Jinkyoo Park
Earlier studies have tried to resolve this issue by using hierarchical MARL to decompose the main task into subproblems or employing an intrinsic reward to induce interactions for learning an effective policy.
Multi-agent Reinforcement Learning reinforcement-learning +1
no code implementations • 22 Jun 2021 • Michael Poli, Stefano Massaroli, Clayton M. Rabideau, Junyoung Park, Atsushi Yamashita, Hajime Asama, Jinkyoo Park
We introduce the framework of continuous-depth graph neural networks (GNNs).
no code implementations • NeurIPS 2021 • Michael Poli, Stefano Massaroli, Luca Scimeca, Seong Joon Oh, Sanghyuk Chun, Atsushi Yamashita, Hajime Asama, Jinkyoo Park, Animesh Garg
Effective control and prediction of dynamical systems often require appropriate handling of continuous-time and discrete, event-triggered processes.
no code implementations • NeurIPS 2021 • Stefano Massaroli, Michael Poli, Sho Sonoda, Taji Suzuki, Jinkyoo Park, Atsushi Yamashita, Hajime Asama
We detail a novel class of implicit neural models.
no code implementations • 7 Jun 2021 • Stefano Massaroli, Michael Poli, Stefano Peluchetti, Jinkyoo Park, Atsushi Yamashita, Hajime Asama
We systematically develop a learning-based treatment of stochastic optimal control (SOC), relying on direct optimization of parametric control policies.
no code implementations • 6 Jun 2021 • Junyoung Park, Sanjar Bakhtiyar, Jinkyoo Park
We propose ScheduleNet, a RL-based real-time scheduler, that can solve various types of multi-agent scheduling problems.
1 code implementation • ICLR 2022 • Junyoung Park, Jinhyun Choo, Jinkyoo Park
We propose the convergent graph solver (CGS), a deep learning method that learns iterative mappings to predict the properties of a graph system at its stationary state (fixed point) with guaranteed convergence.
no code implementations • 2 Jun 2021 • Junyoung Park, Jaehyeong Chun, Sang Hun Kim, Youngkook Kim, Jinkyoo Park
In solving the formulated problem, the proposed framework employs a GNN to learn that node features that embed the spatial structure of the JSSP represented as a graph (representation learning) and derive the optimum scheduling policy that maps the embedded node features to the best scheduling action (policy learning).
no code implementations • 18 Jan 2021 • Heechang Ryu, Hayong Shin, Jinkyoo Park
We propose an algorithm that boosts MARL training using the biased action information of other agents based on a friend-or-foe concept.
Multi-agent Reinforcement Learning reinforcement-learning +1
no code implementations • 14 Jan 2021 • Stefano Massaroli, Michael Poli, Federico Califano, Jinkyoo Park, Atsushi Yamashita, Hajime Asama
We introduce optimal energy shaping as an enhancement of classical passivity-based control methods.
no code implementations • 1 Jan 2021 • Inwook Kim, Jinkyoo Park
We train the proposed model using the random mCVRP instance with different numbers of vehicles, customers, and refueling stations.
no code implementations • 1 Jan 2021 • Hyunwook Kang, SEUNGWOO SCHIN, James Morrison, Jinkyoo Park
Experimental results on solving NP-hard MRRC problems (and IMPS in the Appendix) highlight the near-optimality and transferability of the proposed methods.
no code implementations • 1 Jan 2021 • Junyoung Park, Sanzhar Bakhtiyarov, Jinkyoo Park
From the RL perspective, Minmax mTSP raises several significant challenges, such as the cooperation of multiple workers and the need for a well-engineered reward function.
no code implementations • pproximateinference AABI Symposium 2021 • Yohan Jung, Jinkyoo Park
We then propose a scalable learning method to train the HMM-GPSM model using large-scale data having (1) long sequences of state transitions and (2) a large number of time-series observations for each hidden state.
no code implementations • 20 Sep 2020 • Michael Poli, Stefano Massaroli, Atsushi Yamashita, Hajime Asama, Jinkyoo Park
Continuous-depth learning has recently emerged as a novel perspective on deep learning, improving performance in tasks related to dynamical systems and density estimation.
no code implementations • 12 Aug 2020 • Heechang Ryu, Hayong Shin, Jinkyoo Park
To train the MARL model effectively without designing the intrinsic reward, we propose a learning-based exploration strategy to generate the initial states of a game.
1 code implementation • NeurIPS 2020 • Michael Poli, Stefano Massaroli, Atsushi Yamashita, Hajime Asama, Jinkyoo Park
The infinite-depth paradigm pioneered by Neural ODEs has launched a renaissance in the search for novel dynamical system-inspired deep learning primitives; however, their utilization in problems of non-trivial size has often proved impossible due to poor computational scalability.
no code implementations • 12 Jun 2020 • Yohan Jung, Kyungwoo Song, Jinkyoo Park
To improve the training, we propose an approximate Bayesian inference for the SM kernel.
no code implementations • 18 Mar 2020 • Stefano Massaroli, Michael Poli, Michelangelo Bin, Jinkyoo Park, Atsushi Yamashita, Hajime Asama
We introduce a provably stable variant of neural ordinary differential equations (neural ODEs) whose trajectories evolve on an energy functional parametrised by a neural network.
no code implementations • ICLR Workshop DeepDiffEq 2019 • Michael Poli, Stefano Massaroli, Atsushi Yamashita, Hajime Asama, Jinkyoo Park
In this paper we present a general framework for continuous--time gradient descent, often referred to as gradient flow.
no code implementations • ICLR Workshop DeepDiffEq 2019 • Stefano Massaroli, Michael Poli, Sanzhar Bakhtiyarov, Atsushi Yamashita, Hajime Asama, Jinkyoo Park
Action spaces equipped with parameter sets are a common occurrence in reinforcement learning applications.
Hierarchical Reinforcement Learning reinforcement-learning +1
1 code implementation • NeurIPS 2020 • Stefano Massaroli, Michael Poli, Jinkyoo Park, Atsushi Yamashita, Hajime Asama
Continuous deep learning architectures have recently re-emerged as Neural Ordinary Differential Equations (Neural ODEs).
no code implementations • 7 Jan 2020 • Yohan Jung, Jinkyoo Park
This model can effectively model the sequence of time-series data.
1 code implementation • 18 Nov 2019 • Michael Poli, Stefano Massaroli, Junyoung Park, Atsushi Yamashita, Hajime Asama, Jinkyoo Park
We introduce the framework of continuous--depth graph neural networks (GNNs).
no code implementations • pproximateinference AABI Symposium 2019 • Yohan Jung, Jinkyoo Park
We propose a method for Spectral Mixture kernel approximation using the Reparameterized Random Fourier Feature (R-RFF) in the sense of both general parameter and natural parameter view.
no code implementations • 27 Sep 2019 • Heechang Ryu, Hayong Shin, Jinkyoo Park
Most previous studies on multi-agent reinforcement learning focus on deriving decentralized and cooperative policies to maximize a common reward and rarely consider the transferability of trained policies to new tasks.
2 code implementations • 24 Sep 2019 • Michael Poli, Jinkyoo Park, Ilija Ilievski
Finance is a particularly challenging application area for deep learning models due to low noise-to-signal ratio, non-stationarity, and partial observability.
2 code implementations • 6 Sep 2019 • Stefano Massaroli, Michael Poli, Federico Califano, Angela Faragasso, Jinkyoo Park, Atsushi Yamashita, Hajime Asama
Neural networks are discrete entities: subdivided into discrete layers and parametrized by weights which are iteratively optimized via difference equations.
no code implementations • 29 May 2019 • Hyunwook Kang, Taehwan Kwon, Jinkyoo Park, James R. Morrison
In representing the MRRC problem as a sequential decision-making problem, we observe that each state can be represented as an extension of probabilistic graphical models (PGMs), which we refer to as random PGMs.
no code implementations • 22 Oct 2018 • Heechang Ryu, Hayong Shin, Jinkyoo Park
We propose an efficient multi-agent reinforcement learning approach to derive equilibrium strategies for multi-agents who are participating in a Markov game.
Multi-agent Reinforcement Learning Reinforcement Learning (RL)