21 papers · ranked by Valyu relevance
Xingxing Liang, Yang Ma, Yanghe Feng, Zhong Liu
On-policy deep reinforcement learning algorithms have low data utilization and require significant experience for policy improvement. This paper proposes a proximal policy optimization algorithm with prioritized trajectory replay (PTR-PPO) that combines on-policy and off-policy methods to improve sampling efficiency by…
Rui Zhao, Volker Tresp
In Hindsight Experience Replay (HER), a reinforcement learning agent is trained by treating whatever it has achieved as virtual goals. However, in previous work, the experience was replayed at random, without considering which episode might be the most valuable for learning. In this paper, we develop an energy-based…
Patrick Scheffe, Jianye Xu, Bassam Alrifaee
In prioritized planning for vehicles, vehicles plan trajectories in parallel or in sequence. Parallel prioritized planning offers approximately consistent computation time regardless of the number of vehicles but struggles to guarantee collision-free trajectories. Conversely, sequential prioritized planning can…
Patrick Scheffe, Julius Kahle, Bassam Alrifaee
Distributing computations among agents in large networks reduces computational effort in multi-agent path finding (MAPF). One distribution strategy is prioritized planning (PP). In PP, we couple and prioritize interacting agents to achieve a desired behavior across all agents in the network. We characterize the…
Nada Y. AbdelRahman, Wanchen Jiang, Luke T. Coddington, Sheng Gong + 2 more
Animals efficiently learn to navigate their environment. In the laboratory, naive mice explore their environment via highly structured trajectories and can learn to localize new spatial targets in as few as a handful of trials. It is unclear how such efficient learning is possible, since existing computational models…
Hang Ma, Daniel Harabor, Peter J. Stuckey, Jiaoyang Li + 1 more
We study prioritized planning for Multi-Agent Path Finding (MAPF). Existing prioritized MAPF algorithms depend on rule-of-thumb heuristics and random assignment to determine a fixed total priority ordering of all agents a priori. We instead explore the space of all possible partial priority orderings as part of a novel…
Kimberly Almaraz, Tyler Jang, McKenna Lewis, Titan Ngo + 2 more
The ability to prioritize people living with HIV by risk of future transmissions could aid public health officials in optimizing epidemiological intervention. While methods exist to perform such prioritization based on molecular data, their effectiveness and accuracy are poorly understood, and it is unclear how one can…
Authors not listed
Identifying synthesis routes from knowledge graphs poses challenges beyond retrosynthesis, including path–finding artifacts and data issues. We introduce “SynGPS”, a novel algorithm that overcomes these limitations by identifying viable routes even with common artifacts. SynGPS can resolve nonsensical cycles…
Johannes Smolander, Sini Junttila, Mikko S. Venäläinen, Laura L. Elo
Computational models are needed to infer a representation of the cells, i.e. a trajectory, from single-cell RNA-sequencing data that model cell differentiation during a dynamic process. Although many trajectory inference methods exist, their performance varies greatly depending on the dataset and hence there is a need…
Aida Vatankhah, Ramiro Liscano, Joan García-Haro, Antonio-Javier Garcia-Sanchez + 2 more
The Time-Slotted Channel Hopping (TSCH) protocol is known for its suitability in highly reliable applications within industrial wireless sensor networks. One of the most significant challenges in TSCH is determining a schedule with a minimal slotframe size that can meet the required throughput for a heterogeneous…
Nazanin Zounemat-Kermani, Matthew Richardson, Alen Faiz, Siyao Wang + 12 more
Many longitudinal omics studies contain only a small number of repeated measurements collected before, during, or after an intervention. Existing approaches, including mixed-effects models and generalized additive models, estimate temporal effects but do not generally provide a discrete representation of trajectory…
Varun Gupta, Durg Singh Chauhan, Kamlesh Dutta
The software development process is a complex process, especially when the software has to be released in a phased manner. The high stakes involved and several constraints on resources lead to the selective implementation of the user requirements at each stage of the development. If the requirements considered, do not…
Anwar Ahmed Khan, Sayeed Ghani, Shama Siddiqui
Prioritizing the heterogeneous traffic for Wireless Sensor Networks (WSNs) imposes an important performance challenge for Internet of Things (IoT) applications. Most past preemptive MAC schemes are based on scheduling the high priority packets earlier than those of lower priority. However, in a majority of these…
Carolin Schonard, Tobias Heed, Christian Seegelke
Visuospatial attention is a prerequisite for the performance of visually guided movements: Perceptual discrimination is regularly enhanced at target locations prior to movement initiation. It is known that this attentional prioritization evolves over the time of movement preparation; however, it is not clear whether…
Ali Paikan, Daniele E. Domenichelli, Lorenzo Natale
Real-Time communication is important in distributed applications when timing constraints on task execution and data processing play a fundamental role. Software engineering does not yet specify how real-time properties should be integrated into a publish/subscribe middleware. This article describes an approach for…
Christopher Tyson, Santosh Gaire, Ian Pegg, Abhijit Sarkar
We present a method for tracking densely clustered, high-velocity, indistinguishable objects being spawned at a high rate and moving in a directed force field using only object centroids as inputs and no other image information. The algorithm places minimal restrictions on the velocities or accelerations of the objects…
Audrey Pollien, Evaristo Villaseco Arribas, David Lauvergnat, Federica Agostini
We present an analysis of the performance of the coupled-trajectory schemes for nonadiabatic dynamics derived from the exact factorization of the electron-nuclear wavefunction and implemented in the G-CTMQC code. These algorithms can be seen as variations of the standard Ehrenfest method and Tully surface hopping…
Francesco Talotta, David Lauvergnat, Federica Agostini
The exact factorization of the electron-nuclear wavefunction is applied to the study of the photo- isomerization of a retinal chromophore model. We describe such an ultrafast nonadiabatic process by analyzing the time-dependent potentials of the theory and by mimicking nuclear dynamics with quantum and coupled…
Carlotta Pieroni, Eduarda Sangiogo Gil, Lea-Maria Ibele, Maurizio Persico + 2 more
In this work, we present the first implementation of the coupled-trajectory Tully surface hopping (CT-TSH) suitable for applications to molecular systems. We combine CT-TSH with the semiempirical Floating Occupation Molecular Orbitals-Configuration Interaction (FOMO-CI) electronic structure method to investigate the…
Mohammad Dabbagh, Sai Peck Lee
Due to the budgetary deadlines and time to market constraints, it is essential to prioritize software requirements. The outcome of requirements prioritization is an ordering of requirements which need to be considered first during the software development process. To achieve a high quality software system, both…
Authors not listed
Inverse molecular design aims to generate novel chemical structures that satisfy multiple property constraints, yet reinforcement-learning (RL) fine-tuning can be sensitive to how objectives are converted into a scalar reward. Here, we systematically analyze how scalarization choices and stabilization mechanisms shape…