GRASP is a brand new gradient-based planner for realized dynamics (a “world model”) that makes long-horizon planning sensible by (1) lifting the trajectory into digital states so optimization is parallel throughout time, (2) including stochasticity on to the state iterates for exploration, and (3) reshaping gradients so actions get clear alerts whereas we keep away from brittle “state-input” gradients by way of high-dimensional imaginative and prescient fashions.
Giant, realized world fashions have gotten more and more succesful. They will predict lengthy sequences of future observations in high-dimensional visible areas and generalize throughout duties in ways in which have been troublesome to think about a couple of years in the past. As these fashions scale, they begin to look much less like task-specific predictors and extra like general-purpose simulators.
However having a robust predictive mannequin isn’t the identical as with the ability to use it successfully for management/studying/planning. In follow, long-horizon planning with fashionable world fashions stays fragile: optimization turns into ill-conditioned, non-greedy construction creates dangerous native minima, and high-dimensional latent areas introduce delicate failure modes.
On this weblog put up, I describe the issues that motivated this venture and our strategy to handle them: why planning with fashionable world fashions could be surprisingly fragile, why lengthy horizons are the actual stress take a look at, and what we modified to make gradient-based planning far more sturdy.
This weblog put up discusses work completed with Mike Rabbat, Aditi Krishnapriyan, Yann LeCun, and Amir Bar (* denotes equal advisorship), the place we suggest GRASP.
What’s a world mannequin?
Lately, the time period “world model” is kind of overloaded, and relying on the context can both imply an express dynamics mannequin or some implicit, dependable inside state {that a} generative mannequin depends on (e.g. when an LLM generates chess strikes, whether or not there may be some inside illustration of the board). We give our unfastened working definition beneath.
Suppose you are taking actions $a_t in mathcal{A}$ and observe states $s_t in mathcal{S}$ (photographs, latent vectors, proprioception). A world mannequin is a realized mannequin that, given the present state and a sequence of future actions, predicts what is going to occur subsequent. Formally, it defines a predictive distribution on a sequence of noticed states $s_{t-h:t}$ and present motion $a_t$:
[P_theta(s_{t+1} mid s_{t-h:t},; a_t)]
that approximates the atmosphere’s true conditional $P(s_{t+1} mid s_{t-h:t},; a_t)$. For this weblog put up, we’ll assume a Markovian mannequin $P(s_{t+1} mid s_{t-h:t},; a_t)$ for simplicity (all outcomes right here could be prolonged to the extra normal case), and when the mannequin is deterministic it reduces to a map over states:
[s_{t+1} = F_theta(s_t, a_t).]
In follow the state $s_t$ is commonly a realized latent illustration (e.g., encoded from pixels), so the mannequin operates in a (theoretically) compact, differentiable house. The important thing level is {that a} world mannequin offers you a differentiable simulator; you’ll be able to roll it ahead underneath hypothetical motion sequences and backpropagate by way of the predictions.
Planning: selecting actions by optimizing by way of the mannequin
Given a begin $s_0$ and a purpose $g$, the best planner chooses an motion sequence $mathbf{a}=(a_0,dots,a_{T-1})$ by rolling out the mannequin and minimizing terminal error:
[min_{mathbf{a}} ; | s_T(mathbf{a}) – g |_2^2, quad text{where } s_T(mathbf{a}) = mathcal{F}_{theta}^{T}(s_0,mathbf{a}).]
Right here we use $mathcal{F}^T$ as shorthand for the total rollout by way of the world mannequin (dependence on mannequin parameters $theta$ is implicit):
[mathcal{F}_{theta}^{T}(s_0, mathbf{a}) = F_theta(F_theta(cdots F_theta(s_0, a_0), cdots, a_{T-2}), a_{T-1}).]
Briefly horizons and low-dimensional methods, this could work fairly properly. However as horizons develop and fashions grow to be bigger and extra expressive, its weaknesses grow to be amplified.
So why doesn’t this simply work at scale?
Why long-horizon planning is tough (even when every thing is differentiable)
There are two separate ache factors for the extra normal world mannequin, plus a 3rd that’s particular to realized, deep learning-based fashions.
1) Lengthy-horizon rollouts create deep, ill-conditioned computation graphs
These conversant in backprop by way of time (BPTT) could discover that we’re differentiating by way of a mannequin utilized to itself repeatedly, which can result in the exploding/vanishing gradients downside. Specifically, if we take derivatives (be aware we’re differentiating vector-valued features, leading to Jacobians that we denote with $D_x (cdots)$) with respect to earlier actions (e.g. $a_0$):
[D_{a_0} mathcal{F}_{theta}^{T}(s_0, mathbf{a}) = Bigl(prod_{t=1}^T D_s F_theta(s_t, a_t)Bigr) D_{a_0}F_theta(s_0, a_0).]
We see that the Jacobian’s conditioning scales exponentially with time $T$:
[sigma_{text{max/min}}(D_{a_0}mathcal{F}_{theta}^{T}) sim sigma_{text{max/min}}(D_s F_theta)^{T-1},]
resulting in exploding or vanishing gradients.
2) The panorama is non-greedy and filled with traps
At brief horizons, the grasping answer, the place we transfer straight towards the purpose at each step, is commonly adequate. If you happen to solely have to plan a couple of steps forward, the optimum trajectory normally doesn’t deviate a lot from “head toward $g$” at every step.
As horizons develop, two issues occur. First, longer duties usually tend to require non-greedy conduct: going round a wall, repositioning earlier than pushing, backing as much as take a greater path. And as horizons develop, extra of those non-greedy steps are usually wanted. Second, the optimization house itself scales with horizon: $mathrm{dim}(mathcal{A} occasions cdots occasions mathcal{A}) = Tmathrm{dim}(mathcal{A})$, additional increasing the house of native minima for the optimization downside.
An extended-horizon repair: lifting the dynamics constraint
Suppose we deal with the dynamics constraint $s_{t+1} = F_{theta}(s_t, a_t)$ as a smooth constraint, and we as an alternative optimize the next penalty operate over each actions $(a_0,ldots,a_{T-1})$ and states $(s_0,ldots,s_T)$:
[min_{mathbf{s},mathbf{a}} mathcal{L}(mathbf{s}, mathbf{a}) = sum_{t=0}^{T-1} big|F_theta(s_t,a_t) – s_{t+1}big|_2^2,
quad text{with } s_0 text{ fixed and } s_T=g.]
That is additionally typically referred to as collocation in planning/robotics literature. Observe the lifted formulation shares the identical international minimizers as the unique rollout goal (each are zero precisely when the trajectory is dynamically possible). However the optimization landscapes are very completely different, and we get two rapid advantages:
- Every world mannequin analysis $F_{theta}(s_t,a_t)$ relies upon solely on native variables, so all $T$ phrases could be computed in parallel throughout time, leading to an enormous speed-up for longer horizons, and
- You not backpropagate by way of a single deep $T$-step composition to get a studying sign, because the earlier product of Jacobians now splits right into a sum, e.g.:
[D_{a_0} mathcal{L} = 2(F_theta(s_0, a_0) – s_1).]
Having the ability to optimize states immediately additionally helps with exploration, as we are able to briefly navigate by way of unphysical domains to seek out the optimum plan:
Nevertheless, lunch isn’t free. And certainly, particularly for deep learning-based world fashions, there’s a important situation that makes the above optimization fairly troublesome in follow.
A difficulty for deep learning-based world fashions: sensitivity of state-input gradients
The tl;dr of this part is: immediately optimizing states by way of a deep learning-based $F_{theta}$ is extremely brittle, à la adversarial robustness. Even in case you prepare your world mannequin in a lower-dimensional state house, the coaching course of for the world mannequin makes unseen state landscapes very sharp, whether or not it’s an unseen state itself or just a traditional/orthogonal route to the info manifold.
Adversarial robustness and the “dimpled manifold” mannequin
Adversarial robustness initially checked out classification fashions $f_theta : mathbb{R}^{wtimes h occasions c} to mathbb{R}^Ok$, and confirmed that by following the gradient of a selected logit $nabla f_theta^ok$ from a base picture $x$ (not of sophistication $ok$), you didn’t have to maneuver far alongside $x’ = x + epsilonnabla f_theta^ok$ to make $f_theta$ classify $x’$ as $ok$ (Szegedy et al., 2014; Goodfellow et al., 2015):
Later work has painted a geometrical image for what’s happening: for information close to a low-dimensional manifold $mathcal{M}$, the coaching course of controls conduct in tangential instructions, however doesn’t regularize conduct in orthogonal instructions, thus resulting in delicate conduct (Stutz et al., 2019). One other approach said: $f_theta$ has an affordable Lipschitz fixed when contemplating solely tangential instructions to the info manifold $mathcal{M}$, however can have very excessive Lipschitz constants in regular instructions. The truth is, it typically advantages the mannequin to be sharper in these regular instructions, so it may well match extra sophisticated features extra exactly.
Consequently, such adversarial examples are extremely widespread even for a single given mannequin. Additional, this isn’t simply a pc imaginative and prescient phenomenon; adversarial examples additionally seem in LLMs (Wallace et al., 2019) and in RL (Gleave et al., 2019).
Whereas there are strategies to coach for extra adversarially sturdy fashions, there’s a identified trade-off between mannequin efficiency and adversarial robustness (Tsipras et al., 2019): particularly within the presence of many weakly-correlated variables, the mannequin should be sharper to realize increased efficiency. Certainly, most fashionable coaching algorithms, whether or not in pc imaginative and prescient or LLMs, don’t prepare adversarial robustness out. Thus, a minimum of till deep studying sees a serious regime change, this can be a downside we’re caught with.
Why is adversarial robustness a problem for world mannequin planning?
Contemplate a single element of the dynamics loss we’re optimizing within the lifted state strategy:
[min_{s_t, a_t, s_{t+1}} |F_theta(s_t, a_t) – s_{t+1}|_2^2]
Let’s additional deal with simply the bottom state:
[min_{s_t} |F_theta(s_t, a_t) – s_{t+1}|_2^2.]
Since world fashions are usually skilled on state/motion trajectories $(s_1, a_1, s_2, a_2, ldots)$, the state-data manifold for $F_{theta}$ has dimensionality bounded by the motion house:
[mathrm{dim}(mathcal{M}_s) le mathrm{dim}(mathcal{A}) + 1 + mathrm{dim}(mathcal{R}),]
the place $mathcal{R}$ is a few optionally available house of augmentations (e.g. translations/rotations). Thus, we are able to usually count on $mathrm{dim}(mathcal{M}_s)$ to be a lot decrease than $mathrm{dim}(mathcal{S})$, and thus: it is vitally simple to seek out adversarial examples that hack any state to another desired state.
Consequently, the dynamics optimization
[sum_{t=0}^{T-1} big|F_theta(s_t,a_t) – s_{t+1}big|_2^2]
feels extremely “sticky,” as the bottom factors $s_t$ can simply trick $F_{theta}$ into considering it’s already made its native purpose.1
1. This adversarial robustness situation, whereas notably dangerous for lifted-state approaches, isn’t distinctive to them. Even for serial optimization strategies that optimize by way of the total rollout map $mathcal{F}^T$, it’s attainable to get into unseen states, the place it is vitally simple to have a traditional element fed into the delicate regular elements of $D_s F_{theta}$. The motion Jacobian’s chain rule growth is
[Bigl(prod_{t=1}^T D_s F_theta(s_t, a_t)Bigr) D_{a_0}F_theta(s_0, a_0).]
See what occurs if any stage of the product has any element regular to the info manifold. ↩
Our repair
That is the place our new planner GRASP is available in. The principle remark: whereas $D_s F_{theta}$ is untrustworthy and adversarial, the motion house is normally low-dimensional and exhaustively skilled, so $D_a F_{theta}$ is definitely affordable to optimize by way of and doesn’t undergo from the adversarial robustness situation!
At its core, GRASP builds a first-order lifted state / collocation-based planner that’s solely depending on motion Jacobians by way of the world mannequin. We thus exploit the differentiability of realized world fashions $F_{theta}$, whereas not falling sufferer to the inherent sensitivity of the state Jacobians $D_s F_{theta}$.
GRASP: Gradient RelAxed Stochastic Planner
As famous earlier than, we begin with the collocation planning goal, the place we elevate the states and loosen up dynamics right into a penalty:
[min_{mathbf{s},mathbf{a}} mathcal{L}(mathbf{s}, mathbf{a}) = sum_{t=0}^{T-1} big|F_theta(s_t,a_t) – s_{t+1}big|_2^2,
quad text{with } s_0 text{ fixed and } s_T=g.]
We then make two key additions.
Ingredient 1: Exploration by noising the state iterates
Even with a smoother goal, planning is nonconvex. We introduce exploration by injecting Gaussian noise into the digital state updates throughout optimization.
A easy model:
[s_t leftarrow s_t – eta_s nabla_{s_t}mathcal{L} + sigma_{text{state}} xi, qquad xisimmathcal{N}(0,I).]
Actions are nonetheless up to date by non-stochastic descent:
[a_t leftarrow a_t – eta_a nabla_{a_t}mathcal{L}.]
The state noise helps you “hop” between basins within the lifted house, whereas the actions stay guided by gradients. We discovered that particularly noising states right here (versus actions) finds a great stability of exploration and the flexibility to seek out sharper minima.2
2. As a result of we solely noise the states (and never the actions), the corresponding dynamics should not actually Langevin dynamics. ↩
Ingredient 2: Reshape gradients: cease brittle state-input gradients, hold motion gradients
As mentioned, the delicate pathway is the gradient that flows into the state enter of the world mannequin, (D_s F_{theta}). Essentially the most easy approach to do that initially is to only cease state gradients into (F_{theta}) immediately:
- Let $bar{s}_t$ be the identical worth as $s_t$, however with gradients stopped.
Outline the stop-gradient dynamics loss:
[mathcal{L}_{text{dyn}}^{text{sg}}(mathbf{s},mathbf{a})
= sum_{t=0}^{T-1} big|F_theta(bar{s}_t, a_t) – s_{t+1}big|_2^2.]
This alone doesn’t work. Discover now states solely comply with the earlier state’s step, with out something forcing the bottom states to chase the following ones. Consequently, there are trivial minima for simply stopping on the origin, then just for the ultimate motion making an attempt to get to the purpose in a single step.
Dense purpose shaping
We are able to view the above situation because the purpose’s sign being minimize off fully from earlier states. One method to repair that is to easily add a dense purpose time period all through prediction:
[mathcal{L}_{text{goal}}^{text{sg}}(mathbf{s},mathbf{a})
= sum_{t=0}^{T-1} big|F_theta(bar{s}_t, a_t) – gbig|_2^2.]
In regular settings this might over-bias in direction of the grasping answer of straight chasing the purpose, however that is balanced in our setting by the stop-gradient dynamics loss’s bias in direction of possible dynamics. The ultimate goal is then as follows:
[mathcal{L}(mathbf{s},mathbf{a}) = mathcal{L}_{text{dyn}}^{text{sg}}(mathbf{s},mathbf{a}) + gamma , mathcal{L}_{text{goal}}^{text{sg}}(mathbf{s},mathbf{a}).]
The result’s a planning optimization goal that doesn’t have dependence on state gradients.
Periodic “sync”: briefly return to true rollout gradients
The lifted stop-gradient goal is nice for quick, guided exploration, nevertheless it’s nonetheless an approximation of the unique serial rollout goal.
So each $K_{textual content{sync}}$ iterations, GRASP does a brief refinement part:
- Roll out from $s_0$ utilizing present actions $mathbf{a}$, and take a couple of small gradient steps on the unique serial loss:
[mathbf{a} leftarrow mathbf{a} – eta_{text{sync}},nabla_{mathbf{a}},|s_T(mathbf{a})-g|_2^2.]
The lifted-state optimization nonetheless supplies the core of the optimization, whereas this refinement step provides some help to maintain states and actions grounded in direction of actual trajectories. This refinement step can after all get replaced with a serial planner of your selection (e.g. CEM); the core concept is to nonetheless get a few of the good thing about the full-path synchronization of serial planners, whereas nonetheless principally utilizing the advantages of the lifted-state planning.
How GRASP addresses long-range planning
Collocation-based planners supply a pure repair for long-horizon planning, however this optimization is kind of troublesome by way of fashionable world fashions as a result of adversarial robustness points. GRASP proposes a easy answer for a smoother collocation-based planner, alongside steady stochasticity for exploration. Consequently, longer-horizon planning finally ends up not solely succeeding extra, but additionally discovering such successes sooner:
| Horizon | CEM | GD | LatCo | GRASP |
|---|---|---|---|---|
| H=40 | 61.4% / 35.3s | 51.0% / 18.0s | 15.0% / 598.0s | 59.0% / 8.5s |
| H=50 | 30.2% / 96.2s | 37.6% / 76.3s | 4.2% / 1114.7s | 43.4% / 15.2s |
| H=60 | 7.2% / 83.1s | 16.4% / 146.5s | 2.0% / 231.5s | 26.2% / 49.1s |
| H=70 | 7.8% / 156.1s | 12.0% / 103.1s | 0.0% / — | 16.0% / 79.9s |
| H=80 | 2.8% / 132.2s | 6.4% / 161.3s | 0.0% / — | 10.4% / 58.9s |
Push-T outcomes. Success fee (%) / median time to success. Daring = finest in row. Observe the median success time will bias increased with increased success fee; GRASP manages to be sooner regardless of increased success fee.
What’s subsequent?
There’s nonetheless loads of work to be completed for contemporary world mannequin planners. We wish to exploit the gradient construction of realized world fashions, and collocation (lifted-state optimization) is a pure strategy for long-horizon planning, nevertheless it’s essential to know typical gradient construction right here: clean and informative motion gradients and brittle state gradients. We view GRASP as an preliminary iteration for such planners.
Extension to diffusion-based world fashions (deeper latent timesteps could be seen as smoothed variations of the world mannequin itself), extra refined optimizers and noising methods, and integrating GRASP into both a closed-loop system or RL coverage studying for adaptive long-horizon planning are all pure and attention-grabbing subsequent steps.
I do genuinely assume it’s an thrilling time to be engaged on world mannequin planners. It’s a humorous candy spot the place the background literature (planning and management general) is extremely mature and well-developed, however the present setting (pure planning optimization over fashionable, large-scale world fashions) continues to be closely underexplored. However, as soon as we work out all the suitable concepts, world mannequin planners will probably grow to be as commonplace as RL.
For extra particulars, learn the total paper or go to the venture web site.
Quotation
@article{psenka2026grasp,
title={Parallel Stochastic Gradient-Primarily based Planning for World Fashions},
writer={Michael Psenka and Michael Rabbat and Aditi Krishnapriyan and Yann LeCun and Amir Bar},
12 months={2026},
eprint={2602.00475},
archivePrefix={arXiv},
primaryClass={cs.LG},
url={
}
