diff --git a/docs/static/images/banner.svg b/docs/static/images/banner.svg new file mode 100644 index 0000000..38d501e --- /dev/null +++ b/docs/static/images/banner.svg @@ -0,0 +1,246 @@ + + + + + + + + + + + + + + + + + + + + + + + + + + + + 1. The Vulnerability + + + + + Cost of Information from First Principles + P ~ π(τ) + p = reservation price + M = P - p + + + + + + + + p + + + E[P] + + + + average information rent + COI := E[P] - p + + + + + Why COI Erodes with Agent Saturation + p(1) = min(p1, ..., pN) + P(p(1) > t) = [1 - F(t)]N + + + + + + F(t) + [1 - F(t)]^N + + + + N=1 + + + N=4 + + + N=16 + + As independent query count grows, + realizable markup collapses. + + + + + + + 2. The Behavioral Signals + + + + + From Session Paths to Transition Kernels + + human: start → view → detail → cart → purchase + agent: start → view → detail → view → detail + + + P̂(s'|s) = N(s,s') / Σ N(s,k) + + + + + + transition counts N(s,s') + normalized kernel T + + + + + + 0.00 8.00 0.00 0.00 + 0.00 2.00 5.00 1.00 + 0.00 3.00 2.00 4.00 + 0.00 1.00 0.00 6.00 + + + + + + + + + + 0.00 1.00 0.00 0.00 + 0.00 0.25 0.62 0.13 + 0.00 0.33 0.22 0.45 + 0.00 0.14 0.00 0.86 + + + Kernel shape is the compact behavioral signature used downstream. + + + + + Separability into a Control Signal + ΔH = DKL(T̂' || T̄H) + ΔA = DKL(T̂' || T̄A) + g = ΔH - ΔA + + + + + g = ΔH - ΔA + + + + human + + + + agent + + + + decision boundary + + + g_obs + + Positive gap shifts score toward agent traffic. + + + + + + + + 3. Robust Control & Contamination + + + + + Contamination Generator G(α) + + + + labeled human sessions + + + synthetic agent sessions + + + + + + + + mixed batch for training + + + alpha = 0.33 + + + + human share (1-α) + agent share (α) + + + + + Distributionally Robust Control Layer + + π* = arg maxπ minQ ∈ Uε + Ed ~ Q + [ R(p,d) - λ COIleak(p,τ') ] + + + + + + + + + Uε + + + + P̂N + + + worst-case Q* + + + + + + + + + + inner min picks Q* + + + + + sample demand from Q* + + + + + outer max updates policy + + + Reward is evaluated on demand drawn from Q*, then used for the policy step. + + + diff --git a/docs/static/videos/BehaviorKernelConstructionScene.mp4 b/docs/static/videos/BehaviorKernelConstructionScene.mp4 new file mode 100644 index 0000000..f5ad8db Binary files /dev/null and b/docs/static/videos/BehaviorKernelConstructionScene.mp4 differ diff --git a/docs/static/videos/COIFirstPrinciplesScene.mp4 b/docs/static/videos/COIFirstPrinciplesScene.mp4 new file mode 100644 index 0000000..6a7438b Binary files /dev/null and b/docs/static/videos/COIFirstPrinciplesScene.mp4 differ diff --git a/docs/static/videos/COIOrderStatisticProofScene.mp4 b/docs/static/videos/COIOrderStatisticProofScene.mp4 new file mode 100644 index 0000000..ae20b79 Binary files /dev/null and b/docs/static/videos/COIOrderStatisticProofScene.mp4 differ diff --git a/docs/static/videos/CardMarketAnalogyScene.mp4 b/docs/static/videos/CardMarketAnalogyScene.mp4 new file mode 100644 index 0000000..ce61563 Binary files /dev/null and b/docs/static/videos/CardMarketAnalogyScene.mp4 differ diff --git a/docs/static/videos/ContaminationGeneratorScene.mp4 b/docs/static/videos/ContaminationGeneratorScene.mp4 new file mode 100644 index 0000000..b79a916 Binary files /dev/null and b/docs/static/videos/ContaminationGeneratorScene.mp4 differ diff --git a/docs/static/videos/DefenseOpening.mp4 b/docs/static/videos/DefenseOpening.mp4 new file mode 100644 index 0000000..41ec4ba Binary files /dev/null and b/docs/static/videos/DefenseOpening.mp4 differ diff --git a/docs/static/videos/ObjectiveAndResultsScene.mp4 b/docs/static/videos/ObjectiveAndResultsScene.mp4 new file mode 100644 index 0000000..b400b5f Binary files /dev/null and b/docs/static/videos/ObjectiveAndResultsScene.mp4 differ diff --git a/docs/static/videos/RobustControlScene.mp4 b/docs/static/videos/RobustControlScene.mp4 new file mode 100644 index 0000000..d4450fc Binary files /dev/null and b/docs/static/videos/RobustControlScene.mp4 differ diff --git a/docs/static/videos/SeparabilitySignalScene.mp4 b/docs/static/videos/SeparabilitySignalScene.mp4 new file mode 100644 index 0000000..bae74f1 Binary files /dev/null and b/docs/static/videos/SeparabilitySignalScene.mp4 differ diff --git a/docs/static/videos/SystemLoopScene.mp4 b/docs/static/videos/SystemLoopScene.mp4 new file mode 100644 index 0000000..e901943 Binary files /dev/null and b/docs/static/videos/SystemLoopScene.mp4 differ diff --git a/docs/static/videos/TakeawayScene.mp4 b/docs/static/videos/TakeawayScene.mp4 new file mode 100644 index 0000000..052a954 Binary files /dev/null and b/docs/static/videos/TakeawayScene.mp4 differ diff --git a/paper/defense/manim/scenes.py b/paper/defense/manim/scenes.py index 2ebac50..6a74998 100644 --- a/paper/defense/manim/scenes.py +++ b/paper/defense/manim/scenes.py @@ -1397,6 +1397,171 @@ class TakeawayScene(Scene): self.wait(1.0) +class ThesisBannerPosterScene(Scene): + def construct(self) -> None: + title = Text("PHANTOM", font_size=72, weight="BOLD", color=INK).to_edge(UP) + subtitle = Text( + "Pricing Heuristics Against Non-human Transaction Orchestration", + font_size=24, + color=GREY_B, + ).next_to(title, DOWN, buff=0.05) + + coi_axes = Axes( + x_range=[0, 1, 0.2], + y_range=[0, 1, 0.2], + x_length=3.15, + y_length=1.75, + tips=False, + axis_config={"stroke_width": 1.8, "color": AXIS_INK}, + ) + coi_n1 = coi_axes.plot( + lambda x: (1 - x) ** 1, + x_range=[0, 1], + color=BLUE_D, + stroke_width=4, + ) + coi_n8 = coi_axes.plot( + lambda x: (1 - x) ** 8, + x_range=[0, 1], + color=ORANGE, + stroke_width=4, + ) + coi_hint = Text( + "Order-statistic tail compresses as query count grows", font_size=15 + ) + coi_hint.set_color(GREY_B).next_to(coi_axes, DOWN, buff=0.06) + coi_title = Text("1) COI erosion", font_size=23, weight="BOLD", color=ORANGE) + coi_body = VGroup(coi_axes, coi_n1, coi_n8, coi_hint) + coi_group = VGroup(coi_title, coi_body).arrange(DOWN, buff=0.08) + coi_frame = SurroundingRectangle(coi_group, color=ORANGE, buff=0.14) + coi_frame.set_fill(color=ORANGE, opacity=0.05) + coi_panel = VGroup(coi_frame, coi_group) + + gap_axes = Axes( + x_range=[-8, 8, 2], + y_range=[0.0, 0.2, 0.05], + x_length=3.15, + y_length=1.75, + tips=False, + axis_config={"stroke_width": 1.8, "color": AXIS_INK}, + ) + gap_h = gap_axes.plot( + lambda x: normal_pdf(x, -3.35, 2.67), + x_range=[-8, 8], + color=BLUE_D, + stroke_width=4, + ) + gap_a = gap_axes.plot( + lambda x: normal_pdf(x, 1.65, 2.83), + x_range=[-8, 8], + color=RED_C, + stroke_width=4, + ) + gap_boundary = DashedLine( + gap_axes.c2p(0, 0), + gap_axes.c2p(0, 0.17), + color=GREY_B, + stroke_width=2, + ) + gap_hint = Text( + "Gap score g = Delta_H - Delta_A drives alpha-hat", font_size=15 + ) + gap_hint.set_color(GREY_B).next_to(gap_axes, DOWN, buff=0.06) + gap_title = Text( + "2) Behavioral separability", font_size=23, weight="BOLD", color=GREEN_C + ) + gap_body = VGroup(gap_axes, gap_h, gap_a, gap_boundary, gap_hint) + gap_group = VGroup(gap_title, gap_body).arrange(DOWN, buff=0.08) + gap_frame = SurroundingRectangle(gap_group, color=GREEN_C, buff=0.14) + gap_frame.set_fill(color=GREEN_C, opacity=0.05) + gap_panel = VGroup(gap_frame, gap_group) + + ctrl_title = Text( + "3) Robust pricing control", font_size=23, weight="BOLD", color=HIGHLIGHT + ) + ctrl_signal = MathTex(r"\hat\alpha(\tau')=\sigma(\beta g)", font_size=31) + ctrl_policy = MathTex( + r"\pi^*=\arg\max_\pi\min_{Q\in\mathcal U_\epsilon}\mathbb E[r]", + font_size=29, + color=HIGHLIGHT, + ) + ctrl_steps = VGroup( + card( + "estimate contamination from behavior", + color=GREEN_C, + width=4.0, + height=0.72, + font_size=16, + ), + card( + "optimize price policy under uncertainty", + color=ORANGE, + width=4.0, + height=0.72, + font_size=16, + ), + ).arrange(DOWN, buff=0.18) + ctrl_arrow = Arrow( + ctrl_steps[0].get_bottom(), + ctrl_steps[1].get_top(), + buff=0.06, + color=AXIS_INK, + stroke_width=3, + ) + ctrl_body = VGroup(ctrl_signal, ctrl_policy, ctrl_steps, ctrl_arrow).arrange( + DOWN, buff=0.14 + ) + ctrl_group = VGroup(ctrl_title, ctrl_body).arrange(DOWN, buff=0.08) + ctrl_frame = SurroundingRectangle(ctrl_group, color=HIGHLIGHT, buff=0.14) + ctrl_frame.set_fill(color=HIGHLIGHT, opacity=0.05) + ctrl_panel = VGroup(ctrl_frame, ctrl_group) + + panels = VGroup(coi_panel, gap_panel, ctrl_panel).arrange(RIGHT, buff=0.3) + panels.scale(0.92).next_to(subtitle, DOWN, buff=0.28) + + web = card("web sessions", color=BLUE_D, width=2.2, height=0.7, font_size=17) + kafka = card( + "quote + event logs", color=HIGHLIGHT, width=2.6, height=0.7, font_size=17 + ) + kernel = card( + "transition kernels", color=GREEN_C, width=2.5, height=0.7, font_size=17 + ) + policy = card( + "robust policy", color=ORANGE, width=2.2, height=0.7, font_size=17 + ) + flow_nodes = VGroup(web, kafka, kernel, policy).arrange(RIGHT, buff=0.22) + flow_nodes.to_edge(DOWN, buff=0.52) + flow_arrows = VGroup( + Arrow(web.get_right(), kafka.get_left(), buff=0.05, stroke_width=2.8), + Arrow(kafka.get_right(), kernel.get_left(), buff=0.05, stroke_width=2.8), + Arrow(kernel.get_right(), policy.get_left(), buff=0.05, stroke_width=2.8), + ) + + status = VGroup( + Text("Mann-Whitney p = 0.0006", font_size=19, color=GREEN_C), + Text("Pairwise robust wins: 13/40 objective, 16/40 revenue", font_size=19), + ).arrange(DOWN, buff=0.06) + status[1].set_color(GREY_B) + status.next_to(flow_nodes, UP, buff=0.15) + + footer = Text( + "From mechanism failure to an implementable defense loop", + font_size=25, + color=HIGHLIGHT, + ).next_to(flow_nodes, DOWN, buff=0.13) + + self.add( + title, + subtitle, + panels, + flow_nodes, + flow_arrows, + status, + footer, + ) + self.wait(0.1) + + SCENE_ORDER = [ "DefenseOpening", "CardMarketAnalogyScene", @@ -1410,3 +1575,7 @@ SCENE_ORDER = [ "ObjectiveAndResultsScene", "TakeawayScene", ] + +POSTER_SCENES = ["ThesisBannerPosterScene"] + +AVAILABLE_SCENES = SCENE_ORDER + POSTER_SCENES