TempoNet: Slack-Quantized Transformer-Guided Reinforcement Scheduler for Adaptive Deadline-Centric Real-Time Dispatchs

Rong Fu; Yibo Meng; Zeyu Zhang; Ziming Guo; Jia Yee Tan; Xiaojing Du; Simon James Fong

cs.LGFeb 20, 2026

TempoNet: Slack-Quantized Transformer-Guided Reinforcement Scheduler for Adaptive Deadline-Centric Real-Time Dispatchs

Authors: Rong Fu, Yibo Meng, Zeyu Zhang, Ziming Guo, Jia Yee Tan, Xiaojing Du, Simon James Fong

Organizations: Independent Researcher

Abstract

Real-time schedulers must reason about tight deadlines under strict compute budgets. We present TempoNet, a reinforcement learning scheduler that pairs a permutation-invariant Transformer with a deep Q-approximation. An Urgency Tokenizer discretizes temporal slack into learnable embeddings, stabilizing value learning and capturing deadline proximity. A latency-aware sparse attention stack with blockwise top-k selection and locality-sensitive chunking enables global reasoning over unordered task sets with near-linear scaling and sub-millisecond inference. A multicore mapping layer converts contextualized Q-scores into processor assignments through masked-greedy selection or differentiable matching. Extensive evaluations on industrial mixed-criticality traces and large multiprocessor settings show consistent gains in deadline fulfillment over analytic schedulers and neural baselines, together with improved optimization stability. Diagnostics include sensitivity analyses for slack quantization, attention-driven policy interpretation, hardware-in-the-loop and kernel micro-benchmarks, and robustness under stress with simple runtime mitigations; we also report sample-efficiency benefits from behavioral-cloning pretraining and compatibility with an actor-critic variant without altering the inference pipeline. These results establish a practical framework for Transformer-based decision making in high-throughput real-time scheduling.

Figures & tables

Figure 1: Overview of the TempoNet architecture for adaptive deadline-centric real-time dispatching. The pipeline initiates with the Urgency Tokenizer (UT) , which transforms continuous per-job slack $s_{i}(t)$ into a discrete vocabulary via Slack Quantization (clip and floor) and retrieves learned Urgency Tokens $\mathbf{x}_{i}(t)$ from an embedding matrix $\mathbf{E}$ . These tokens are gathered into a Token Assembly matrix $\mathbf{X}(t)$ , maintaining permutation invariance. At the core, a Transformer Encoder stacks $L$ blocks of Multi-Head Attention and Position-wise Feed-Forward Networks to generate contextualized task representations $\mathbf{H}^{(L)}$ . The Q-Value Projection layer maps these representations to per-token Q-scores $\mathbf{q}(t)$ , which are then passed through a Multicore Mapping module that utilizes an Iterative Masked-Greedy or bipartite matching strategy to determine the final action $a_{t}$ . The framework is optimized via a Deep Q-Learning loop, where experiences are stored in a Replay Buffer $\mathcal{D}$ to update the primary network $Q_{\theta}$ against a soft-updated Target Network $Q_{\theta^{-}}$ .

Figure 2: Attention-Criticality Correlation Analysis

Figure 3: Attention Focus Distribution Across Tasks heatmap

Figure 4: Computational Time Scaling with System Size

Figure 5: Entropy Distribution Across Transformer Layers

Figure 6: Distribution of Q-Values Across Utilization Levels

Scheduling Approach	Mean	Median	Std. Dev.	Min	Max
PPO [ 30 ]	0.68	0.82	0.31	0.00	0.98
A3C [ 31 ]	0.71	0.84	0.29	0.01	0.99
FF-DQN	0.74	0.86	0.26	0.00	1.00
Rainbow DQN [ 32 ]	0.78	0.87	0.25	0.00	1.00
Offline RL [ 33 ]	0.79	0.84	0.27	0.01	0.98
GraSP-RL [ 34 ]	0.80	0.85	0.23	0.02	0.99

Table 1: Comparative Deadline Compliance at

u\approx 0.87

(200 Randomized Tasksets)

Heads	Hit Rate	Std Dev	$\Delta$ Gain
2	80.3%	0.31	-5.5%
4	85.0 %	0.27	0.0%
6	84.7%	0.33	-0.3%
8	84.9%	0.35	-0.1%

Table 2: Attention Head Impact (

L=2

d=128

)

Figure 7: Embedding Dimension Performance–Computation Tradeoff

Layers	Hit Rate	Latency	$\Delta$ Hit
1	76.2%	0.42ms	-8.8%
2	85.0 %	0.51ms	0.0%
3	86.1%	0.71ms	+1.1%
4	85.7%	0.94ms	+0.7%

Table 3: Depth Impact on Performance (

H=4

d=128

)

Method	Tasks	Success Rate	Time (s)
MHQISSO (EDF)	100	97.8%	20.4
DRL-Based [ 41 ]	100	97.5%	9.0
LSTM-PPO [ 42 ]	100	97.6%	7.5
ENF-S [ 43 ]	100	97.8%	8.0
PSO-Based [ 44 ]	100	97.0%	10.2
Transformer-based [ 11 ]	100	97.9%	8.7

Table 4: Large-scale Scheduling Efficiency

Method	PITMD (%)	ART	Time (s)
DIOS	87.28	16.72	–
FCFS	9.83	21.20	–
EDF [ 45 ]	20.81	20.68	–
Mo-QIGA [ 46 ]	83.21	15.11	0.48
HQIGA [ 46 ]	85.70	16.34	0.52
Transformer-based [ 35 ]	88.00	14.20	0.45

Table 5: Industrial Scenario Performance Metrics

Methodology	Compliance Rate	Improvement
Rate Monotonic (RM)	11.67%	–
Earliest Deadline First (EDF)	11.67%	–
Feedforward DQN (FF-DQN)	71.43%	–
TempoNet (Proposed)	79.00%	+7.57%

Table 6: Deadline compliance rates on the standard task configuration.

Appendix figures & tables19 assets

Supplementary material from the paper’s appendix.

Appendix

Figure 8: Miss rate comparison between EDF empirical, TempoNet, and the utilization-based reference $1-1/U$ . The utilization-based curve is shown only for empirical reference and does not imply a per-instance theoretical lower bound for the distributions evaluated.

Variant	Input type	Hit Rate (%)	$\Delta$ vs TempoNet	Training $\sigma^{2}$
FF-DQN-cont	scalar slack (raw)	$74.8\pm 2.9$	$-12.2$	$4.5\times 10^{-3}$
FF-DQN-norm	scalar slack (z-score)	$76.1\pm 2.4$	$-10.9$	$3.9\times 10^{-3}$
FF-DQN-MLP	scalar slack + 2-layer MLP	$79.5\pm 2.0$	$-7.5$	$3.1\times 10^{-3}$
TempoNet	quantised + embedding	$87.0\pm 1.9$	—	$1.7\times 10^{-3}$
TempoNet w/o UT	continuous slack (concat)	$81.3\pm 2.2$	$-5.7$	$2.4\times 10^{-3}$

Appendix

Table 7: Deadline-compliance (Hit Rate) for TempoNet and continuous-slack baselines on 200 task-sets (utilisation

0.6

–

1.0

). Results are presented as mean

\pm

std over 5 seeds. “

\Delta

vs TempoNet” shows the difference in percentage points relative to TempoNet.

Method	Pattern	Latency @600 (ms)	Memory traffic	Scheduling-specific?
Explicit Sparse Transformer [ 18 ]	static column-drop	$0.68$	$1.00\times$	No
Efficient Sparse Attention [ 49 ]	learned block-drop	$0.65$	$0.95\times$	No
TempoNet	deadline-aware block + chunk top- $k$	$0.42$	$0.62\times$	Yes

Appendix

Table 8: Per-layer attention kernel comparison (median latency at

N=600

, and relative DRAM traffic).

Quartile	Mean Deadline (ms)	Meet Rate (%)
Q1 (shortest)	$12\pm 3$	$86.8\pm 2.1$
Q2	$25\pm 4$	$87.2\pm 1.9$
Q3	$55\pm 9$	$86.5\pm 2.3$
Q4 (longest)	$180\pm 35$	$85.9\pm 2.7$
KS-test $p$ -value	0.18 (no bias)

Appendix

Table 9: Deadline-meet rate across quartiles of absolute deadline under Pareto-distributed deadlines (heavy-tailed). KS-test

p

-value indicates no statistically significant bias toward short tasks.

Method	Avg. Response Time (ms)	Deadline Meet Rate (%)
SRPT (preemptive, optimal mean) [ 50 ]	$14.2\pm 0.8$	$68.3\pm 2.1$
TempoNet (ours)	$\mathbf{12.4\pm 1.0}$	$\mathbf{87.0\pm 1.9}$

Appendix

Table 10: Head-to-head comparison between SRPT (optimal for mean response time) and TempoNet on 200 heterogeneous task-sets. TempoNet wins on both mean response time and deadline compliance.

Mapping	PITMD (%)	ART (ms)	Inference (µs)	Comment
A: masked-greedy	90.1	12.4	420	default, sub-ms
B: Sinkhorn	90.6	12.1	860	+0.5 pp, $\times$ 2 latency

Appendix

Table 11: Multicore-mapping trade-off on 600-task industrial trace (8 cores).

Reward	Compliance (%)	95th lateness (ms)	Train stability $\sigma$	Energy index
Binary	89.2	18.3	0.27	1.18
R1 (lateness penalty)	89.1	13.1	0.29	1.15
R2 (early bonus)	89.5	15.0	0.26	1.21
R3 (slack-sensitive)	90.4	11.7	0.23	1.09
R4 (risk-aware)	90.1	12.4	0.24	1.12
R5 (energy-aware)	88.7	14.2	0.25	1.00

Appendix

Table 12: Extended reward-shaping study on a 600-task trace. Energy is normalised to the minimal value observed under the energy-aware scheme.

Task-set	Encoder (µs)	Mapping (µs)	End-to-end (µs)	Max N@1 ms
64	82 $\pm$ 5	18 $\pm$ 2	105 $\pm$ 7	$\sim$ 1,050
200	145 $\pm$ 8	35 $\pm$ 3	185 $\pm$ 10	$\sim$ 720
600	298 $\pm$ 12	71 $\pm$ 5	375 $\pm$ 17	$\sim$ 430

Appendix

Table 13: Hardware-in-the-loop latency breakdown and 1 ms real-time bound (median

\pm

MAD).

Figure 9: Real-time boundary: synthetic inference latency versus number of active tasks $N$ with a 1 ms budget line. Vertical dashed line marks the interpolated upper bound $N_{\max}$ . Replace with measured timings for an exact per-system bound.

Adaptation	0-shot	5-ep	10-ep	20-ep	Oracle
Compliance (%)	84.1 $\pm$ 1.2	87.3 $\pm$ 0.8	88.9 $\pm$ 0.5	89.2 $\pm$ 0.4	90.0 $\pm$ 0.3
Oracle gap (%)	6.6	3.0	1.2	0.9	0.0

Appendix

Table 14: Non-stationary robustness: zero-shot vs few-shot adaptation (200 tasks, 8 cores).

$Q$	Binning	Hit rate (%)	ART (ms)	Train $\sigma^{2}$ ( $\times 10^{-3}$ )
8	uniform	83.5 $\pm$ 1.1	13.8	3.9
32	uniform	86.4 $\pm$ 0.9	12.9	2.2
128	uniform	87.0 $\pm$ 0.8	12.4	1.7
128	log-spaced	87.2 $\pm$ 0.7	12.3	1.6
128	data-driven*	87.3 $\pm$ 0.6	12.2	1.5
*K-means on empirical slack distribution, $K=Q$ .

Appendix

Table 15: Slack-quantisation sensitivity (200 task-sets, 8 cores).

Figure 10: Hyperparameter sensitivity: filled contour of synthetic deadline compliance versus slack bin width $\Delta$ and embedding dimension $d$ . Illustrative; replace with measured performance grid for final submission.

Cores	Util	Zero-shot (%)	5-ep (%)	Oracle (%)	Gap (%)
4	0.6	88.9	89.4	89.6	0.7
4	0.9	85.1	86.8	87.2	2.1
16	0.6	89.3	89.7	90.0	0.7
16	0.9	86.0	87.5	88.1	2.1
32	1.1	82.7	85.9	87.4	4.7

Appendix

Table 16: Zero-shot and few-shot core-count transfer (trained on 8-core @ util 0.8).

Dense full attention	$\mathcal{O}(N^{2})$ . This is the worst-case cost when every query attends to all keys.
Block Top-k with fixed block size and budget	$\Theta(N)$ . Holds when $B=O(1)$ and $k=O(1)$ .
Block Top-k with $B=\Theta(\sqrt{N}),k=\Theta(\sqrt{N})$	$\Theta\!\bigl(N^{1.5}/\sqrt{\log N}\bigr)$ . Average-case analytic estimate under chunking.
Measured (optimized CUDA kernel, finite $N$ range)	$\Theta(N^{1.1})$ (empirical fit). Observed when kernel fusion and symmetric sparsity reduce constants.

Appendix

Table 17: Per-layer attention complexity under different regimes.

Figure 11: Comparative convergence of standard $\epsilon$ -greedy and the enhanced uncertainty-based exploration on heterogeneous workloads. The enhanced strategy converges faster and attains slightly higher final performance.

Initialization	Episodes to 85% compliance	Final compliance (%)	Wall-clock hours
Random initialization	$92\pm 8$	$89.2\pm 0.4$	$4.2$
BC warm-start (pilot)	$\mathbf{51\pm 5}$	$\mathbf{89.5\pm 0.3}$	$\mathbf{2.3}$

Appendix

Table 18: Sample-efficiency comparison: BC pre-training versus random initialization on a 600-task industrial trace with 8 cores.

Method	Compliance (%)	95th lateness (ms)	Peak memory (MB)
TempoNet	$71.3\pm 2.1$	$28.7\pm 3.4$	$312\pm 8$
GNN-based RL [ 51 ]	$68.9\pm 2.7$	$31.2\pm 4.1$	$295\pm 10$
EDF [ 47 ]	$52.1\pm 3.0$	$45.6\pm 5.2$	—
SRPT [ 50 ]	$63.5\pm 2.5$	$35.1\pm 3.9$	—

Appendix

Table 19: Adversarial stress-test results. Compliance denotes fraction of tasks meeting deadline; 95th lateness is the 95th percentile of lateness in milliseconds; memory usage is peak resident memory in megabytes.

Figure 12: Probability of selection conditioned on quantized slack and normalized remaining execution.

Pair	Action agreement	Interpretation
TempoNet vs. EDF	68%	leans toward shorter jobs relative to EDF
TempoNet vs. SRPT	71%	gives additional weight to deadlines vs. SRPT
TempoNet vs. distilled rule	91%	closely matched by a single weighted rule

Appendix

Table 20: Action-sequence agreement between TempoNet and classical schedulers or the distilled linear rule.

Real-Time Scheduling

Real-time scheduling allocates processors, accelerators, and network resources so tasks meet timing deadlines under workload and resource constraints.

See all 30 papers

A Deep Reinforcement Learning (DRL)-Based Transformer Method for Solving the Open Shop Scheduling Problem
Mar 25, 2026Faezeh Ardali, Mwembezi A. Nyelele, Gerald M. KnappCombinatorial Optimization Neural Combinatorial Optimization
Louisiana State University, Baton Rouge, LA, USA · University of Minnesota Duluth, Duluth, MN, USA
RED: Adaptive Real-Time DAG Scheduling for Robotic Inference under Environmental Dynamics
May 21, 2026Zexin Li, Tao Ren, Johnathan Liu +2Efficient Neural Network Inference Real-Time Scheduling
University of California, Riverside, USA · University of Pittsburgh, USA · University of Maryland, Baltimore County, USA +1
Harmonizing Real-Time Constraints and Long-Horizon Reasoning: An Asynchronous Agentic Framework for Dynamic Scheduling
May 28, 2026Shijie Cao, Yuan Yuan, Jing LiuLanguage Model-Based Control Real-Time Scheduling
School of Computer Science and Engineering, Beihang University, Beijing 100191, China · Shenzhen Loop Area Institute, Shenzhen, China · Qingdao Research Institute, Beihang University +3

Mar 25, 2026J/K move · Enter open · S save

A Deep Reinforcement Learning (DRL)-Based Transformer Method for Solving the Open Shop Scheduling Problem

Faezeh Ardali, Mwembezi A. Nyelele, Gerald M. Knapp

Louisiana State University, Baton Rouge, LA, USA · University of Minnesota Duluth, Duluth, MN, USA

Combinatorial Optimization Neural Combinatorial Optimization RL for Combinatorial Optimization Transformer-Based RL

Open paper AskPDF arXiv

The open shop scheduling problem (OSSP) arises in many industrial and service settings but remains computationally challenging as the number of jobs and machines increases. While exact methods quickly become intractable, classical dispatching rules and metaheuristics may require substantial tuning to maintain solution quality at large scales. This study develops a Transformer-based scheduling policy for OSSP using an encoder-decoder architecture with multi-head attention. The model is trained on Taillard benchmark instances (4x4, 5x5, 7x7, and 10x10) using only the processing-time matrix as input and produces feasible schedules with makespans typically within 15-30% of best-known values. To evaluate scalability, the trained policy is applied without retraining to randomly generated instances from 40x40 to 100x100 and compared against classical dispatching heuristics, including SPT, LPT, MWKR, and EST. Across these large instances, the Transformer achieved average gaps of 12.89-15.12% relative to a standard lower bound. Compared with EST, the Transformer remained competitive, typically within a modest margin, while substantially outperforming SPT and LPT. These results indicate that a Transformer policy trained on small OSSP instances can generalize to substantially larger problems and provide a feature-light, learning-based alternative to classical dispatching rules.

Abstract

Real-Time Scheduling

Explore similar work

A Deep Reinforcement Learning (DRL)-Based Transformer Method for Solving the Open Shop Scheduling Problem

RED: Adaptive Real-Time DAG Scheduling for Robotic Inference under Environmental Dynamics

Harmonizing Real-Time Constraints and Long-Horizon Reasoning: An Asynchronous Agentic Framework for Dynamic Scheduling